IQ Bell Curve Explained: Mean, SD and Percentiles

The IQ bell curve is the single most useful idea for making sense of an intelligence score. When a large, representative group of people takes a well-designed test, their results cluster around a central value and thin out steadily toward the extremes. Plotted on a graph, those results trace a smooth, symmetrical shape that statisticians call a normal distribution and everyone else calls a bell curve. Learning to read the iq bell curve tells you far more than a lone number can: it reveals where a score sits relative to everyone else, why some results are common while others are genuinely rare, and why any IQ figure only means something in relation to a reference group. This guide walks through the mean, the standard deviation, the famous 68-95-99.7 rule, and percentiles; for a broader tour of the numbers, see our overview of the IQ scale.

What the IQ bell curve is

A distribution is simply a picture of how often each possible outcome shows up in a population. Height, shoe size, blood pressure and reaction time all tend to follow a bell-shaped pattern: most people land near the middle, and fewer and fewer appear as you move toward either edge. Cognitive ability, as measured by a broad test, behaves the same way. The bell curve is not something the numbers happen to fall into by luck. Modern IQ scores are deliberately designed to fit it. Test makers collect responses from thousands of people, then convert raw answers into a standardized scale so that the results form a normal distribution with known properties. That design choice is what makes scores comparable from one test to another and from one person to the next. Because the curve is symmetrical, there are as many people below the centre as above it, and a score of 100 sits exactly in the middle by definition.

Mean 100 and standard deviation 15

Two numbers define the entire curve. The first is the mean, set at 100. This is the average score and the peak of the bell, the most common outcome. The second is the standard deviation, set at 15 on most modern scales such as the Wechsler tests. The standard deviation measures spread: how tightly or loosely scores bunch around the average. A small standard deviation would make a tall, narrow curve; a large one flattens and widens it. With a standard deviation of 15, a score of 115 is exactly one standard deviation above the mean, 130 is two above, and 145 is three above. Going the other way, 85 is one standard deviation below and 70 is two below. This is why those particular numbers appear so often in discussions of IQ; they are the tidy milestones of the scale. For context on what counts as typical, see what a normal IQ is and how IQ is calculated. It is worth noting that not every test uses 15; some older or specialised scales use 16 or 24, which is why the same ability can produce slightly different numbers on different tests.

The 68-95-99.7 rule

The normal distribution follows a reliable pattern often called the 68-95-99.7 rule, or the empirical rule. About 68 percent of people score within one standard deviation of the mean, meaning between 85 and 115. About 95 percent fall within two standard deviations, between 70 and 130. And about 99.7 percent land within three standard deviations, between 55 and 145. Turn those figures around and the rarity of extreme scores becomes clear. If 95 percent of people sit between 70 and 130, then only 5 percent lie outside that band, split evenly between the two tails. That leaves roughly 2.5 percent above 130 and about 2.5 percent below 70. Push out to three standard deviations and only about 0.15 percent of people exceed 145. These proportions are not arbitrary; they follow directly from the mathematics of the curve, which is why they hold across well-constructed tests rather than depending on any single one.

From scores to percentiles

A percentile translates a raw score into plain language: it tells you the percentage of people who scored at or below your level. A score of 100 sits at the 50th percentile, because half of the population scores lower. A score of 115, one standard deviation up, corresponds to roughly the 84th percentile, meaning about 84 percent of people score at or below it. A score of 130 reaches about the 98th percentile, and 145 sits near the 99.9th. Percentiles are often more intuitive than the raw number because they answer the question people really care about: how do I compare with everyone else? They also make the shape of the curve visible. Near the middle, a few points of IQ move you across many percentile ranks, because that is where most people are packed together. Out in the tails, the same few points barely shift your percentile, because so few people are there to overtake. If you want to explore this step by step, our guide on how to read your percentile breaks the process down.

Why very high scores are rare

The bell curve explains something that often surprises people: why the gap between 100 and 115 feels ordinary while the gap between 130 and 145 feels enormous. Both spans are exactly one standard deviation wide, yet the number of people involved is wildly different. Moving from the 50th to the 84th percentile covers a third of the population. Moving from the 98th to the 99.9th percentile covers barely one person in fifty. The tail thins out fast. This is also why claims of extraordinarily high scores should be read with care. A figure like 160 or 180 sits so far into the tail that only a handful of people in a million would reach it, and at that extreme the measurement itself becomes shaky, because there are too few test-takers to calibrate the scale reliably. For what a strong but attainable score looks like, see what an IQ of 130 means. Rarity is a mathematical feature of the distribution, not a special badge; a very high score is uncommon precisely because the curve leaves so little room out there.

Scores are always relative

Perhaps the most important lesson of the bell curve is that an IQ score is relative, never absolute. Your number does not measure a fixed quantity of intelligence the way a ruler measures length. It measures where you land compared with a reference population, usually people of a similar age. The test is standardized on that group, the average of that group is anchored at 100, and your score describes your position within it. Change the reference group and the same performance can yield a different number. This is why scores are periodically re-normed, and why the well-documented rise in raw scores over generations, known as the Flynn effect, is corrected for by resetting the average back to 100. It is also why an online result should be read as a useful indication of where you sit, not as a clinical measurement or a fixed label. A score from an online test like IQTesta gives you a reasonable estimate of your standing on the curve, and you can take a free IQ test to see where you land; a formal assessment of cognitive ability is a separate, professional process. Understanding the bell curve keeps any single number in perspective: informative, but only ever a snapshot relative to others.