This free hypergeometric distribution calculator finds the probability of getting exactly x successes, at most x, at least x or a value between two counts, when you draw a sample without replacement from a finite population. It shows the step-by-step working with your numbers, a bar chart with the outcomes you asked for highlighted, the mean, variance, standard deviation, mode and median, a full probability table and the exact TI-84 nCr keystrokes. The TI-84, TI-84 Plus CE and TI-83 Plus have no built-in hypgeompdf command, so the calculator writes the combination formula the way Texas Instruments documents it. A second tab finds the smallest sample size that gives you a target chance of at least r successes.
Contents
How to Use the Hypergeometric Distribution Calculator
- Enter the population. N is the total number of items. K is how many of those count as a success (red cards, defectives, seniors in a class).
- Enter the sample size n, the number of items you draw. The draws are without replacement: once an item is picked, it cannot be picked again.
- Pick the probability you need: P(X = x), P(X ≤ x), P(X < x), P(X ≥ x), P(X > x) or P(a ≤ X ≤ b), and type x (and b).
- Read the answer in the purple box. The table under it lists all five probabilities for the same x with the matching TI-84 nCr command, so you can answer a whole homework question at once.
- Open the extras: step-by-step working, TI-84 keystrokes for your numbers and the full probability table. The Draws needed tab finds the smallest n that gives you a target chance of at least r successes.
Results update as you type. Copy share link saves your inputs in the URL. If each draw were independent with the same p, you would use the binomial distribution calculator instead. The combinations in the formula are the same C(n, r) as on the permutation and combination calculator.
What Is the Hypergeometric Distribution?
The hypergeometric distribution describes the number of successes in a sample drawn without replacement from a finite group. Deal 5 cards from a 52-card deck and count the hearts, pull 4 phones from a box of 20 that contains 5 defectives, or pick a committee of 3 from a class of 10 that has 4 seniors: the count of “successes†in the sample is a hypergeometric random variable.
Four conditions make a setting hypergeometric:
- A finite population of N items, of which K are classified as successes.
- A fixed sample size n, decided before you start drawing.
- Sampling without replacement: each draw changes what is left, so the probability of success is not the same on every draw.
- X counts successes in the sample, not the draw on which the first one appears (that would be a different model).
We write X ~ Hypergeometric(N, K, n). The possible values of X run from max(0, n − (N − K)) to min(n, K): you cannot draw more successes than exist, and if the sample is large you may be forced to draw some.
Hypergeometric Distribution Formula
P(X = x) = C(K, x) × C(N − K, n − x) / C(N, n)
P(X ≤ x) = Σ P(X = k) from k = kmin to x μ = n × K / N
σ² = n × (K/N) × ((N − K)/N) × ((N − n)/(N − 1)) σ = √σ²
C(a, b) means “a choose bâ€, the number of ways to pick b items from a. The numerator chooses the x successes from the K available and the remaining n − x draws from the failures; the denominator is the number of ways to choose the whole sample from N. On a TI-84, C(a, b) is a nCr b under MATH PRB. The last factor in the variance, (N − n)/(N − 1), is the finite-population correction: because you do not put items back, X is less spread out than a binomial with the same p = K/N.
Worked Example: 10 Items, 4 Successes, Draw 3
The calculator opens with this question: A box holds 10 light bulbs, 4 of which are long-life. You pick 3 at random. What is the probability that exactly 2 of them are long-life? So N = 10, K = 4, n = 3 and x = 2.
- Exactly 2: P(X = 2) = C(4, 2) × C(6, 1) / C(10, 3) = 6 × 6 / 120 = 36/120 = 0.3.
- At most 2: P(X ≤ 2) = P(X = 0) + P(X = 1) + P(X = 2) = 20/120 + 60/120 + 36/120 = 116/120 = 0.9667.
- Fewer than 2: P(X < 2) = P(X ≤ 1) = 80/120 = 0.6667.
- At least 2: P(X ≥ 2) = 1 − P(X ≤ 1) = 40/120 = 0.3333.
- More than 2: P(X > 2) = P(X = 3) = C(4, 3) × C(6, 0) / 120 = 4/120 = 0.0333.
The mean is 3 × 4 / 10 = 1.2 long-life bulbs, the variance is 0.56 and the standard deviation is √0.56 ≈ 0.7483. The most likely value (the mode) and the median are both 1. On the TI-84, type (4 nCr 2)(6 nCr 1)/(10 nCr 3) and press ENTER to get .3. The Example button loads a classic cards question: 5 cards from a 52-card deck, P(exactly 3 hearts) = C(13, 3) × C(39, 2) / C(52, 5) ≈ 0.0815.
At Least, More Than and Between
Cumulative questions are just sums of the same formula. “At most x†adds every possible value up to x. “Fewer than x†is the same as “at most x − 1â€. “At least x†and “more than x†are easier as complements:
P(X ≥ x) = 1 − P(X ≤ x − 1) P(X > x) = 1 − P(X ≤ x)
P(a ≤ X ≤ b) = P(X ≤ b) − P(X ≤ a − 1)
The results table lists all five single-x probabilities at once, each with the TI-84 command. For a range, the calculator subtracts two cumulative sums so you do not have to add every term by hand.
Mean, Variance and the Finite Population
The mean n × K / N is the sample size times the success rate in the population: if 40% of the box is long-life and you draw 3, you expect 1.2 long-life bulbs. The variance looks like the binomial variance n p (1 − p) multiplied by (N − n)/(N − 1). When the sample is a tiny slice of a large population that factor is almost 1 and hypergeometric and binomial almost agree. When you draw a large share of the box, the factor shrinks and the counts cluster more tightly around the mean. If you draw the entire population, the variance is 0: X is certain to equal K.
How Many Draws Until You Are 95% Sure?
A common planning question is the reverse: how many items should I draw to have a 95% chance of at least one success? There is no short algebra for it. You increase n until P(X ≥ 1) = 1 − C(N − K, n) / C(N, n) clears the target, and you take the smallest such n. With the default box (N = 10, K = 4) that n is 5: P(at least one long-life bulb in 5 draws) = 1 − C(6, 5) / C(10, 5) = 1 − 6/252 ≈ 0.9762, while 4 draws only reach ≈ 0.9286. The Example on that tab uses 50 tickets with 5 winners and a 90% target: you need 18 draws. Both are on the Draws needed tab, with the TI-84 check.
Hypergeometric vs Binomial
| Hypergeometric | Binomial | |
|---|---|---|
| How you draw | Without replacement, from a finite population | Independent trials, or with replacement |
| Success chance | Changes after every draw | The same p on every trial |
| Parameters | N, K, n | n, p |
| TI-84 command | No built-in pdf; use nCr | binompdf and binomcdf under DISTR |
| When they agree | When n is a small slice of N, p ≈ K/N and the finite-population factor is close to 1 | |
Use hypergeometric when the population is listed and items are not put back. Use binomial when each trial is independent (coin flips, a huge population, or sampling with replacement). The geometric distribution is different again: you keep going until the first success, so the number of trials is not fixed. The Poisson calculator is the rare-event model for counts in a fixed interval.
How to Calculate a Hypergeometric Probability on a TI-84
Texas Instruments documents this directly: the TI-84 Plus, TI-84 Plus CE and TI-83 Plus have no hypgeompdf or hypgeomcdf in the DISTR menu. You enter the combination formula on the home screen.
- For the default P(X = 2), you want
(4 nCr 2)(6 nCr 1)/(10 nCr 3). - Type 4, press MATH, arrow right three times to PRB, press 3 for 3:nCr, then type 2.
- Type ( 6, nCr, 1, ) ÷ ( 10, nCr, 3, ), then ENTER. The calculator shows .3.
- For a cumulative probability such as P(X ≤ 2), wrap the general term in a sum:
sum(seq((4 nCr X)((6 nCr (3-X))/(10 nCr 3)),X,0,2)). 2nd STAT (LIST) → MATH 5:sum(, then LIST OPS 5:seq(. The variable is X,T,θ,n. - For “at leastâ€, type 1 − in front of the complementary sum. You can also put x values in L1 and the formula in the L2 header.
Try every keystroke on our TI-84 Plus CE calculator online. The same nCr key is what the nPr and nCr calculator is built around.
Common Mistakes
- Looking for hypgeompdf on the TI-84. It is not there, on the Plus, the Plus CE or the TI-83 Plus. Use MATH PRB nCr, not DISTR.
- Using binomial when items are not replaced. If the population is small, p changes after every draw and binomial overstates the spread.
- Swapping N, K, n and x. N is the population, K the successes in it, n the sample size, x the successes in the sample. A common mix-up is typing x in the K slot.
- Forgetting the possible range. If n = 8 and K = 3, X cannot be 0 when N = 10, because 8 draws from 10 must include at least 1 of the 3 successes.
- Typing “at least†as a single nCr term. P(X ≥ x) is a sum, or 1 minus P(X ≤ x − 1).
Why This Hypergeometric Calculator Is Different
- All six probability types (exactly, at most, fewer than, at least, more than, between), with all five single-x answers and their TI-84 nCr commands in one table.
- Honest TI-84 keystrokes. The calculator does not invent a hypgeompdf command. It writes the nCr formula Texas Instruments tells you to use, including sum(seq( for cumulative questions.
- Draws needed for a target chance of at least r successes, with the check at n and at n − 1.
- A bar chart and a full table with your outcomes highlighted, plus the mean, variance, standard deviation, mode and median.
- Checked against SciPy and exact fractions on more than a thousand cases, including the classic 0.3 textbook example and Texas Instruments’ own N = 12, K = 3, n = 4, x = 1 check (P = 28/55 ≈ 0.5091). It works in dark mode, makes share links and fits a phone screen.
Hypergeometric Distribution Calculator FAQ
TI 84 Calc PRO is not affiliated with or endorsed by Texas Instruments. TI-84 Plus CE is a trademark of Texas Instruments.
About this page