Normal Distribution Explained: The Bell Curve, Z-Scores, and the Empirical Rule

What makes a distribution normal

A normal distribution is a smooth, symmetric, bell-shaped curve described completely by two numbers: its mean, which sets where the peak sits, and its standard deviation, which sets how wide the bell is. The mean, median, and mode all fall at that same central point, and the curve is symmetric around it, so a value is just as likely to land a given distance above the mean as below. The tails stretch out in both directions but get vanishingly thin, so extreme values are possible but rare.

Change the mean and the whole bell slides left or right without changing shape. Change the standard deviation and the bell gets narrower and taller (values tightly clustered) or wider and flatter (values more spread out). Because every normal curve shares this shape, once you can read one of them you can read all of them — which is what the standard normal distribution, covered below, lets you do.

• A normal distribution is symmetric and bell-shaped, defined entirely by its mean and standard deviation.
• The mean sets the center; the standard deviation sets the width.
• Mean, median, and mode coincide at the peak, and the two tails are mirror images.

The empirical rule (68-95-99.7)

The empirical rule is the fastest way to reason about a normal distribution without a calculator. For any normal distribution, about 68% of values fall within one standard deviation of the mean, about 95% within two, and about 99.7% within three. This 68-95-99.7 pattern holds whatever the mean and standard deviation are, because every normal curve has the same proportions.

Suppose adult resting heart rates are roughly normal with a mean of 70 beats per minute and a standard deviation of 8. The rule immediately tells you that about 68% of people fall between 62 and 78, about 95% between 54 and 86, and almost everyone between 46 and 94. A reading of 94 sits three standard deviations above the mean, which the rule flags as genuinely unusual — exactly the kind of quick judgment it is built for.

• About 68% of values lie within 1 standard deviation of the mean, 95% within 2, and 99.7% within 3.
• The rule works for every normal distribution regardless of its mean or standard deviation.
• It gives you an instant sense of which values are typical and which are rare.

Z-scores: putting everything on one scale

A z-score expresses a value as the number of standard deviations it sits from the mean: z = (x − μ) ÷ σ. A z-score of 0 is exactly average, +1.5 is one and a half standard deviations above the mean, and −2 is two standard deviations below. Converting raw values into z-scores is called standardizing, and it strips away the original units.

That is what makes z-scores so useful for comparison. A student who scores 85 on a test with mean 75 and standard deviation 5 has a z-score of +2; a student who scores 90 on a different test with mean 82 and standard deviation 8 has a z-score of +1. Even though 90 is the higher raw score, the first student performed better relative to their group. Standardizing also maps any normal distribution onto the standard normal distribution — mean 0, standard deviation 1 — the reference curve every probability table and calculator function uses.

• z = (value − mean) ÷ standard deviation; it counts standard deviations from the mean.
• Positive z is above average, negative z is below, and z = 0 is exactly average.
• Standardizing removes units so scores from different scales can be compared directly.

Why the normal distribution shows up everywhere

The normal distribution is not just mathematically convenient — it appears throughout nature and measurement. The Central Limit Theorem says that when you add up or average many small, independent influences, the result tends toward a normal distribution even if the individual pieces are not normal. Human height is the sum of many genetic and environmental nudges; measurement error accumulates many tiny random disturbances.

This is also why the normal distribution underpins inference. The average of a sample behaves normally as the sample grows, whatever the shape of the original data, and that single fact is what lets confidence intervals, z-tests, and t-tests use the bell curve to reason about sample means.

• The Central Limit Theorem explains why sums and averages of many small effects turn out normal.
• Many natural measurements — heights, errors — are approximately normal for this reason.
• The normality of sample averages is what powers confidence intervals and hypothesis tests.

Reading probabilities and avoiding common traps

Because area under a normal curve equals probability, a z-score converts directly into the proportion of values below, above, or between points. A z-score of 1.645 sits at the 95th percentile — about 95% of values fall below it — which is why that number turns up in one-sided tests and interval bounds. Percentiles, probabilities, and z-scores are three views of the same curve.

Two cautions keep you honest. First, not all data is normal: strongly skewed data, counts, and bounded measurements often are not, so check a histogram before applying the empirical rule. Second, the tails matter — the normal model treats extreme events as extraordinarily rare, so data with fatter tails throws outliers more often than the model predicts. StatRise's Distribution Explorer lets you slide the mean and standard deviation and watch the density curve change shape, while the Normal CDF calculator turns a cutoff into a probability with each step shown.

• Area under the curve is probability, so a z-score converts directly into a percentile.
• Check for normality with a histogram before applying the empirical rule.
• Real data can have fatter tails than the normal model, making rare events more likely than it predicts.

Frequently asked questions