📚 Trusted by students, educators & researchers since 2017.
If you roll a die 10,000 times, plot the results on a graph, and squint, you will see a shape. If you measure the heights of 10,000 adult men and plot them, you will see a completely different shape.
In statistics, these shapes are called Probability Distributions.
A distribution is simply a mathematical map. It tells you exactly how likely any given outcome is. If you know the shape of your data, you can predict the future.
While there are dozens of distributions in statistics, 90% of your work will involve just four. Here is how to know which shape your data will take.
1. The Uniform Distribution: The Great Equaliser
The Shape: A perfectly flat rectangle. The Rule: Every single outcome has the exact same probability of happening.
Imagine rolling a standard, six-sided die. You have a 1-in-6 chance of rolling a 1, a 1-in-6 chance of rolling a 2, and so on. If you roll that die millions of times and build a bar chart of the results, all six bars will be exactly the same height.
When to use it: Whenever the mechanism creating the data is perfectly random and unbiased.
- Examples: Drawing a card from a shuffled deck, spinning a roulette wheel, or generating a random number on a computer.
2. The Binomial Distribution: The Coin Flipper
The Shape: A bumpy curve that only exists at whole numbers. The Rule: You have a fixed number of trials, and every trial has only two possible outcomes (Success or Failure).
Imagine flipping a coin 10 times and counting how many “Heads” you get. Could you get 0 Heads? Yes, but it’s incredibly rare. Could you get 10 Heads? Yes, also rare. The most likely outcome is 5 Heads.
If you map out the probabilities of getting exactly 0, 1, 2… all the way up to 10 Heads, it forms a distinct mountain shape. Importantly, you cannot flip 5.5 Heads. The data is discrete, meaning the graph is made of distinct, separate bars.
When to use it: When you are counting “wins” out of a set number of attempts.
- Examples: How many free throws a basketball player makes out of 20 attempts. How many defective lightbulbs are found in a batch of 100.
3. The Poisson Distribution: The Waiting Game
The Shape: A mountain that is squished to the left and trails off to the right (right-skewed). The Rule: Counting how many times an event happens over a fixed amount of time or space.
Unlike the Binomial distribution, where you have a set number of “tries,” the Poisson distribution is about waiting.
Imagine counting how many shooting stars you see in an hour. On average, you might see 3. Sometimes you see 2, sometimes you see 4. You rarely see 0. But could you see 20? Yes. It’s incredibly unlikely, but it’s theoretically possible.
The left side of the graph is blocked by zero (you can’t see negative shooting stars), but the right side of the graph stretches out forever into infinity, creating a long “tail.”
When to use it: When predicting the frequency of rare, independent events over time.
- Examples: How many emails arrive in your inbox between 2:00 PM and 3:00 PM. How many car accidents happen at a specific intersection in a month.
4. The Normal Distribution: The Law of the Universe
The Shape: The classic, symmetrical Bell Curve. The Rule: Data is continuous, clustered tightly around an average, and equally likely to deviate above or below that average.
This is the king of all distributions. It appears everywhere in nature because of a mathematical phenomenon called the Central Limit Theorem. When countless small, random factors add up to create a final result, the data forms a perfect bell curve.
Take human height. Your height is determined by genetics, childhood nutrition, sleep, hormones, and hundreds of other tiny variables. Because of all these independent factors clashing together, the vast majority of people end up near the average height. Very few people are exceptionally short, and very few are exceptionally tall. The curve is perfectly symmetrical.
When to use it: When measuring continuous, natural phenomena, or when taking the averages of large samples.
- Examples: Human height, blood pressure, IQ scores, or measurement errors on a factory assembly line.
Use this interactive visualization to see how the mathematical “rules” of your data instantly change the shape of the distribution:
Probability Distributions Explorer
Select a distribution type to see how the mathematical “rules” change the shape of the data.
Uniform Distribution
Every outcome has the exact same probability. Used when events are perfectly random and unbiased.