Normal Distribution — Bell Curve, Z-Score & Formula

A normal distribution is a bell-shaped probability model where values near the mean are most common and values farther away become less common in a symmetric way. If you are trying to understand the bell curve, z-score, or normal distribution formula, the key idea is simple: the mean sets the center, and the standard deviation sets the spread.

This model is useful only when the normal shape is a reasonable fit for the data or situation. When that condition holds, you can estimate typical ranges, compare values with z-scores, and interpret how unusual a result is.

What the bell curve means

If a variable follows a normal distribution, values near the mean are more common than values far away. The left and right sides mirror each other, so being $2$ standard deviations above the mean is just as unusual as being $2$ standard deviations below it.

You will often see the notation

$$X \sim N(\mu, \sigma^2)$$

This means the random variable $X$ is modeled as normal with mean $\mu$ and variance $\sigma^2$. Since variance is $\sigma^2$, the standard deviation is $\sigma$, where $\sigma > 0$.

Normal distribution formula, in plain language

The normal density formula is

$$f(x) = \frac{1}{\sigma \sqrt{2\pi}} e^{-(x-\mu)^2/(2\sigma^2)}$$

You do not need to memorize every part of the formula to use the idea well. What matters most is that $\mu$ moves the curve left or right, while $\sigma$ makes it narrower or wider.

This formula describes density, not the probability of one exact value. For a continuous model, probabilities come from intervals such as $P(X < 80)$ or $P(65 \le X \le 85)$.

How mean, standard deviation, and z-score connect

Changing the mean moves the curve left or right. Changing the standard deviation makes the curve narrower or wider. A small $\sigma$ means values are packed tightly around the mean. A larger $\sigma$ means they are more spread out.

To compare one value with the rest of the distribution, use the z-score:

$$z = \frac{x - \mu}{\sigma}$$

This tells you relative position in standard deviation units. If $z = 1.5$, the value is $1.5$ standard deviations above the mean. If $z = -2$, it is $2$ standard deviations below the mean.

For a normal model, one practical shortcut is the empirical rule:

$$\text{about } 68\% \text{ of values lie within } \mu \pm \sigma$$ $$\text{about } 95\% \text{ of values lie within } \mu \pm 2\sigma$$ $$\text{about } 99.7\% \text{ of values lie within } \mu \pm 3\sigma$$

Use this only when a normal model is actually reasonable. It is a useful approximation, not a guarantee for every real data set.

Worked example with z-score and bell curve

Suppose exam scores are modeled by

$$X \sim N(70, 10^2)$$

So the mean score is $70$ and the standard deviation is $10$.

First, use the empirical rule. About $68\%$ of scores should fall within one standard deviation of the mean:

$$70 \pm 10$$

So the quick interval is

$$60 \text{ to } 80$$

About $95\%$ of scores should fall within two standard deviations:

$$70 \pm 2(10) = 70 \pm 20$$

So that interval is

$$50 \text{ to } 90$$

Now take one student who scored $85$. The z-score is

$$z = \frac{85 - 70}{10} = 1.5$$

That means the score is $1.5$ standard deviations above the mean. This is the fastest useful reading: the score is clearly above average, but not extremely far into the tail.

Common mistakes with normal distribution problems

Treating every bell-shaped graph as normal

Some data are skewed, heavy-tailed, or have multiple peaks. In those cases, a normal model may be a poor fit even if the graph looks roughly rounded.

Confusing density with probability

The formula $f(x)$ is not the probability that $X$ equals one exact number. For continuous distributions, exact-point probability is $0$, so you work with intervals instead.

Using the empirical rule without checking the model

The $68$-$95$-$99.7$ rule belongs to the normal distribution. It should not be applied automatically to any data set.

Mixing up variance and standard deviation

Variance is $\sigma^2$. The z-score uses $\sigma$, not $\sigma^2$.

When the normal distribution is used

The normal distribution appears often when measurements cluster around a central value and extreme values are relatively rare. It is common in measurement error models, test-score interpretation, quality control, and in the study of sample averages.

That does not mean all real data are normal. It means the normal model is a useful approximation when the shape, context, and assumptions make that approximation reasonable.

Try a similar problem

Change the example to $X \sim N(100, 15^2)$ and compute the z-score of $130$. Then find the interval that covers about $95\%$ of values. Trying your own version with a different mean or standard deviation is a good way to see how the bell curve changes.