The Bell Curve

The Gaussian distribution, named after Carl Friedrich Gauss, is the most important probability distribution in science. Its bell-shaped curve appears whenever many small, independent random effects combine: measurement errors, biological variation, test scores, thermal noise. Two parameters define the entire shape. The mean μ sets where the peak sits. The standard deviation σ sets how wide the bell spreads. Everything else follows from these two numbers.

The 68-95-99.7 Rule

For any Gaussian distribution, 68% of values fall within one standard deviation of the mean (μ ± σ), 95% fall within two (μ ± 2σ), and 99.7% fall within three (μ ± 3σ). For the standard normal: 68% between −1 and 1, 95% between −2 and 2. Generate enough values and the smooth curve above will visually confirm these proportions. The colored dots near the center cluster in cool indigo; the rare outliers glow warm amber.

Why the Bell Curve Appears Everywhere

The Central Limit Theorem, one of the most profound results in mathematics, proves that the average of many independent random quantities converges to a Gaussian distribution regardless of how the individual quantities are distributed. Height is the sum of contributions from hundreds of genes and environmental factors. An exam score reflects thousands of micro-decisions during study and testing. A manufacturing tolerance absorbs vibrations, temperature fluctuations, and material variations. Each contributing factor can follow any distribution at all; the combined result gravitates toward the bell curve. This universality is why Gauss's distribution earned its place on the German ten-mark banknote.

The Galton Board

In 1889, Francis Galton built a physical demonstration called the quincunx: a vertical board studded with rows of pegs. A ball dropped from the top bounces left or right at each peg with equal probability. After passing through all rows, balls accumulate in bins at the bottom, forming a bell curve. Each ball's final position is the sum of many independent left-right bounces. The dot scatter on this page works the same way: each generated value is one "ball," and the colored dots accumulate into the bell shape. The more you generate, the smoother and more vivid the curve becomes.

Generating Gaussian Numbers

This tool uses the Box-Muller transform, published by George Box and Mervin Muller in 1958. Two uniform random numbers from crypto.getRandomValues() enter a logarithmic-trigonometric transformation and emerge as a perfectly Gaussian pair. The algorithm is elegant: z = √(−2 ln u₁) · cos(2π u₂), then result = μ + z · σ. The uniform inputs come from hardware-level entropy in your device. The Gaussian output stays in your browser.

In the Classroom

Project the curve canvas on screen and generate 100 values. Students watch the smooth empirical curve rise toward the theoretical outline in real time. Each dot that appears carries a color: indigo near the center, purple in the middle zones, amber at the extremes. The visual immediately communicates that central values are common and outliers are rare. For a concrete scenario, use /gaussian/100/15 (the IQ distribution) or /gaussian/170/7 (adult heights in centimeters). Ask students to predict what fraction of values will land beyond μ + 2σ. The answer (about 2.3%) consistently surprises them.

Private by Architecture

Every Gaussian value on this page is computed inside your browser via the Box-Muller transform. The server delivers the page and educational content. Your device creates every number. Generation history lives in localStorage under your control. The server stores no data, sets no cookies, and never sees your results.