A probability distribution is a map of the future. It tells us all the possible outcomes of a random event and how likely each one is. Think of rolling a standard six-sided die. You know the possible outcomes are 1, 2, 3, 4, 5, or 6. A probability distribution goes a step further and tells you the probability for each of those outcomes.

For a fair die, each outcome has an equal chance of happening. There are six sides, so the probability of rolling any specific number is 1 out of 6, or about 16.7%. We can lay this out in a simple table.

This table is a probability distribution. It’s a formal way of organizing what we intuitively know about rolling a die. The variable that represents the outcome of our experiment, like the number rolled on a die, is called a random variable.

Random variables, and therefore probability distributions, come in two main types: discrete and continuous. The distinction is all about what kind of numbers the outcomes can be.

Our die roll is a perfect example. You can roll a 4 or a 5, but you can't roll a 4.5. Other examples include the number of heads in three coin flips (you can get 0, 1, 2, or 3) or the number of defective items in a batch. The values are distinct and separate.

On the other hand, some data isn't countable.

Think about the height of a person. It could be 1.75 meters, but it could also be 1.751 meters, or 1.7512 meters, depending on how precisely you measure. There are infinite possibilities between any two heights. Other examples are temperature, time, and weight. The distribution for these variables is often shown as a smooth curve, where the area under the curve represents probability.

Simply classifying a distribution as discrete or continuous is just the beginning. To really understand and compare distributions, we use a few key numbers to summarize them. These are like vital statistics for our data.

For a discrete distribution, you calculate the mean by multiplying each possible outcome by its probability, then summing those products.

Let's find the mean for our die roll example:

$E[X] = (1 \cdot \frac{1}{6}) + (2 \cdot \frac{1}{6}) + (3 \cdot \frac{1}{6}) + (4 \cdot \frac{1}{6}) + (5 \cdot \frac{1}{6}) + (6 \cdot \frac{1}{6})$

$E[X] = \frac{1+2+3+4+5+6}{6} = \frac{21}{6} = 3.5$

Even though you can't roll a 3.5, it's the long-term average if you rolled the die thousands of times. It tells us the center of the distribution.

The mean gives us the center, but it doesn't tell us about the spread of the data. Are the outcomes tightly clustered around the mean, or are they all over the place? For that, we need variance and standard deviation.

The formula for variance looks a bit complex, but the idea is simple. For each outcome, you find the difference between it and the mean, square that difference, multiply it by the outcome's probability, and then add them all up.

The squaring is important. It ensures that distances from the mean are always positive, and it gives more weight to outcomes that are farther away.

Variance is measured in squared units (like "squared points" for our die), which is hard to conceptualize. Taking the square root brings it back to the original units, giving us the standard deviation. A low standard deviation means the outcomes are very consistent and predictable. A high standard deviation means they are much more spread out and unpredictable.

These concepts are the building blocks for understanding how data behaves. By knowing the type of distribution and its key properties, we can start to make sense of uncertainty and make better predictions.