Mean, Variance and Standard Deviation of a Binomial Distribution, Explained

These three numbers show up under every result on the calculator, and if you've ever wondered what they're actually telling you beyond "just a formula," here's the plain version.

Mean (μ = n · p)

This is the number of successes you'd expect on average if you repeated the whole experiment a huge number of times. Flip a coin 20 times, and the mean is 10 heads (20 × 0.5). That doesn't mean you'll get exactly 10 heads every time — most runs won't — it just means 10 is the center that your actual results will cluster around.

Quick way to think about it: mean = trials × chance of success. Guess randomly on 40 MCQs with 4 options each, and your expected score is 40 × 0.25 = 10 correct — not because you'll always score exactly 10, but because that's where your typical score lands.

Variance (σ² = n · p · (1 − p))

Variance measures how spread out the results tend to be around that mean. Notice the p(1−p) part — it's biggest when p = 0.5 and shrinks as p gets closer to 0 or 1. That actually makes intuitive sense: if p = 0.5, outcomes are genuinely uncertain and can swing further. If p = 0.98, almost everything succeeds almost every time, so there's very little room for the results to vary.

Standard deviation (σ = √variance)

Standard deviation is just variance brought back down to the original units. Variance for a coin-flip-style experiment ends up in "successes squared," which isn't intuitive to picture — standard deviation fixes that by taking the square root, giving you a number in plain "number of successes" again.

As a rough rule of thumb, most of your actual results will land within about 2 standard deviations of the mean. Flip a coin 100 times: mean = 50, σ ≈ 5. So getting somewhere between 40 and 60 heads is common; getting 70+ would be a genuinely unusual run.

Why this matters beyond homework

These numbers are how you tell "normal variation" apart from "something's actually off." If a machine should defect 2% of parts on average (mean known) but you suddenly measure a result way outside the expected standard deviation range, that's a signal something changed — a broken part, a shift in materials, whatever. Mean tells you what to expect; standard deviation tells you how surprised to be when reality doesn't match it exactly.

See it in action

Every calculation on the calculator shows all three numbers automatically, so you can build intuition just by trying different n and p values and watching how mean and spread respond.