Geometric Distribution
You'll learn: How to model the number of attempts needed to achieve your first success in repeated, independent trials.
What Is the Geometric Distribution?
Imagine you're flipping a coin, trying to get heads. You might get it on your first try, or you might need 2, 3, or even 10 flips. The geometric distribution models exactly this scenario: how many trials do you need until you get your first success?
Unlike the binomial distribution (which counts total successes in a fixed number of trials), the geometric distribution focuses on when that first success happens.
The Setup
You need:
- Independent trials (each attempt doesn't affect others)
- Each trial has the same probability of success,
p - You stop as soon as you get your first success
The Formula
The probability that your first success occurs on trial k is:
P(X = k) = (1 - p)^(k-1) · p
Breaking this down:
(1 - p)^(k-1)means you failedk-1times in a rowpmeans you succeeded on the k-th trial
Example: You're testing a program that crashes randomly with probability 0.2. What's the probability it first crashes on the 3rd run?
P(X = 3) = (1 - 0.2)^(3-1) · 0.2 = (0.8)² · 0.2 = 0.64 · 0.2 = 0.128
You had two successful runs, then crashed on the third.
Key Takeaway: The geometric distribution models "trials until first success" using P(X = k) = (1 - p)^(k-1) · p, where p is the success probability and k is the trial number of the first success.