The Normal: The Bell Everyone Talks About

"The bell curve is one of statistics' superstars — and like every superstar, it gets used in places it doesn't belong."

What the Bell Actually Is

The normal distribution (also called Gaussian, after Carl Friedrich Gauss) is the continuous distribution defined by two numbers: mean μ (where the peak sits) and standard deviation σ (how spread out the bell is). Its probability density looks like the famous bell shape, perfectly symmetric, with the peak at the mean and the spread controlled entirely by σ.

The PDF formula is intimidating-looking but unimportant for the citizen: the visual intuition is what matters. The bell falls off on both sides equally; the rate of fall-off is controlled by σ. Small σ = narrow tall bell. Large σ = wide flat bell.

Why It Shows Up So Often (Preview of Track 03)

The Central Limit Theorem, which we will derive in Track 03, says that when you sum (or average) many small independent random effects, the result tends to look normal — regardless of what the individual effects looked like. This is why measurement errors, adult heights, IQ scores, and a thousand other quantities have roughly bell-shaped distributions.

The bell isn't a mystical universal law of nature. It's the mathematical consequence of one specific setup: many small independent factors adding up. When that setup holds, the bell shows up automatically. When it doesn't hold — when factors are correlated, or one factor dominates, or the underlying process is multiplicative rather than additive — the bell does not show up, and assuming it will lie.

The Famous Rule of Thumb

For a normal distribution, the 68–95–99.7 rule holds:

About 68% of values fall within 1σ of the mean.
About 95% within 2σ.
About 99.7% within 3σ.

This is the lens that Track 04 will sharpen: σ is a unit of intuitive surprise. A value at 2σ is moderately surprising; at 3σ, very surprising; at 5σ or 6σ, effectively impossible under a normal assumption. When 5σ events happen often in your real data, the data isn't normal — full stop.

The Trap to Plant Early

Assuming normality is comfortable but expensive when wrong. The normal has thin tails — extreme values become exponentially rare as you go out. Real-world distributions with fat tails (income, asset returns, network traffic, earthquake magnitudes) have extreme values that are NOT exponentially rare, and any model that fitted a normal to them is going to be blindsided. The bell curve is one of the zoo's animals, not the zoo. Track 07 is where this trap is fully dismantled.

Code

The 68-95-99.7 rule, simulated·python

import numpy as np
rng = np.random.default_rng(8)

# Generate 100,000 samples from a normal with mean 100, std 15 (IQ-like).
x = rng.normal(loc=100, scale=15, size=100_000)

# The 68-95-99.7 rule, checked empirically.
for k in (1, 2, 3, 4, 5):
    within = ((x > 100 - k * 15) & (x < 100 + k * 15)).mean()
    print(f"Within {k} sigma: {within:.4f}   (expected ~{[0.6827, 0.9545, 0.9973, 0.99994, 0.999999][k-1]:.4f})")

# Notice how quickly the tails thin out.
# 5-sigma in a normal: about 6 in a million. In real fat-tailed data it can
# happen more like 6 in ten-thousand. Same number on the page; vastly
# different real-world meaning. Misreading this is the seed of Track 07.

Exercise

Pick three quantities from real life that you suspect are roughly normal (adult shoe size in your country, daily steps in your routine week, time to commute on a typical morning). For each, ask: is this the sum of many small independent factors? If yes, the normal is plausible. If you can name a dominating single factor (one rainy day blows up your commute), be suspicious — the bell may not fit.

Hint

The Central Limit Theorem's precondition is 'many small independent factors.' Check the precondition before trusting the conclusion.

What the Bell Actually Is

Why It Shows Up So Often (Preview of Track 03)

The Famous Rule of Thumb

The Trap to Plant Early

Code

External links

Exercise

Progress

Comments 0