A Super Mega Bowl guide
Flip a coin ten times. How many heads should you expect, and how surprised should you be by any given number? The binomial distribution answers exactly that. It is the math of counting successes across repeated yes-or-no trials, and it is simpler than its name suggests.
You are looking at a binomial distribution whenever four things are true. You run a fixed number of trials. Each trial has just two outcomes, which we can call success and failure. Every trial has the same probability of success. And the trials are independent. Coin flips fit perfectly: ten flips, heads or tails, 50 percent each, no memory between them. So do many real cases, such as how many of 20 manufactured parts are defective, or how many of 100 visitors click a button.
Flip a coin ten times and getting exactly five heads is the single most likely result, but it still happens only about 24.6 percent of the time. The reason is counting. There is only one way to get zero heads, all tails. But there are many different orderings that produce five heads, so five is favored simply because there are more paths to it. The extremes are rare because there are so few ways to reach them.
| Heads in 10 flips | Chance |
|---|---|
| 0 or 10 | about 0.1 percent each |
| 3 or 7 | about 11.7 percent each |
| 4 or 6 | about 20.5 percent each |
| 5 | about 24.6 percent |
The chance of getting exactly k successes in n trials, when each success has probability p, is:
P(k) = C(n, k) × p^k × (1 - p)^(n - k)
It has three pieces. The middle term, p to the power k, is the chance of your k successes. The last term is the chance of the remaining failures. The first term, C(n, k), is the number of different orderings that give k successes, and it is what makes the middle values so much more likely than the extremes. For a fair coin p is 0.5, so this reduces to counting orderings and dividing by the total.
Plot the chances and you get a hump centered on the expected number of successes, which is simply n times p. For 10 coin flips that center is 5, and for 100 flips it is 50. As the number of trials grows, this hump smooths into the familiar bell curve, which is why so many real measurements end up bell shaped. The spread around the center grows with the square root of n, the same square-root behavior behind the law of large numbers.
Anywhere you repeat an independent yes-or-no trial and count the successes, the binomial distribution is quietly at work.
Build the distribution → Flip 10 coins over and over and tally the number of heads. The five-heads bar will pull ahead, and the shape will fill in exactly as predicted.The binomial distribution counts successes across a fixed number of independent yes-or-no trials. Middle results are most likely because more orderings lead to them, the center sits at n times p, and the whole thing grows into a bell curve as trials pile up. Behind a lot of everyday statistics is nothing more exotic than counting coin flips.