Discrete Probability Distribution: From Intuition to Precision
Imagine you're about to roll a fair six-sided die. Before it lands, you know something important: the outcome will be one of six numbers — 1, 2, 3, 4, 5, or 6. You also know that each number is equally likely. That's your first taste of a discrete probability distribution: a complete description of what can happen and how likely each possibility is.
The word "discrete" means separate, countable. The outcomes are like individual points — you can list them. This is different from something like "the height of a randomly chosen student," which can be any value in a continuous range. Here, we're dealing with things you can count: number of heads in three coin tosses, the sum of two dice, the number of customers arriving at a shop in an hour.
The Intuition
A discrete probability distribution answers two questions:
- What are all the possible outcomes? (The sample space)
- What probability does each outcome carry? (The chance it occurs)
The key rule: the probabilities of all possible outcomes must add up to exactly 1. That makes sense — something has to happen, and the total chance of all possibilities is certainty.
Think of a spinner divided into slices. Each slice is an outcome, and the size of the slice is its probability. The whole circle is 1 (or 100%). That's your distribution.
The Precise Statement
Formally, a discrete probability distribution is a function P that assigns a probability to each possible outcome x in a countable set X (the sample space), such that:
- For every outcome x, 0≤P(x)≤1
- The sum over all outcomes is exactly 1: ∑x∈XP(x)=1
The function P is called the probability mass function (PMF). It gives the "mass" or weight of probability at each discrete point.
P(X=x)=p(x),where 0≤p(x)≤1 and ∑xp(x)=1
A Concrete Example
Consider tossing a fair coin twice. Let X be the number of heads.
| Outcome (x) | How it happens | Probability P(x) |
|---|
| 0 | TT | 41 |
| 1 | HT, TH | 42=21 |
| 2 | HH | 41 |
Check: 41+21+41=1. That's a valid discrete probability distribution.
Always verify that probabilities sum to 1. If they don't, you've either missed an outcome or made an arithmetic error.
What It's Used For
Once you have a distribution, you can compute:
- Expected value (the long-run average): E[X]=∑x⋅P(x)
- Variance (how spread out the outcomes are): Var(X)=∑(x−E[X])2⋅P(x)
These are the tools for making predictions and decisions under uncertainty — from gambling to insurance to quality control in manufacturing.
Common Pitfall
A common mistake is to think that if an outcome is possible, its probability must be non-zero. That's true. But the reverse is not: just because a probability is non-zero doesn't mean the outcome is equally likely as others. Always compute, don't assume.
The Big Picture
A discrete probability distribution is your complete map of an uncertain situation where outcomes are countable. It tells you not just what could happen, but exactly how likely each thing is. That's the foundation of all probability and statistics — and it starts with this simple, powerful idea.