tenticle

Teaching Probability

Posted on 2025-10-17 by xkollar in Math, Probability.

Here is a crude representation of what I use as a visual aid for explaining probability: Unit Square.

1 1

Boringly simple, isn’t it?

It has some nice properties. Let’s have an event AA that will happen with probability (A)\mathbb{P}(A).

1 ℙ(A) ℙ(A)

Notice that both the length of the shorter edge of the red region and the area of the red region are equal to (A)\mathbb{P}(A). While nice and useful, this is simultaneously horrible as it stretches your brain to think about probability as a length (one-dimensional) and as an area (two-dimensional), neither of which it actually is, since it is just a ratio (dimensionless). Anyway, if you can stomach that, you can use it to draw pictures that can help you reason about probabilities.

Let’s look at a illustrative example in the following picture.

ℙ(B|A) ℙ(A) ℙ(A&B) ℙ(B) ℙ(A|B) ℙ(A&B)

From that we can easily write down the following two formulas:

(B|A)=(A&B)(A)(A|B)=(A&B)(B) \mathbb{P}(B|A) = \frac{\mathbb{P}(A\&B)}{\mathbb{P}(A)} \quad \mathbb{P}(A|B) = \frac{\mathbb{P}(A\&B)}{\mathbb{P}(B)}

And from there through

(A&B)=(A|B)(B) \mathbb{P}(A\&B) = \mathbb{P}(A|B)\cdot\mathbb{P}(B)

we get straight to

(B|A)=(A|B)(B)(A) \mathbb{P}(B|A) = \frac{\mathbb{P}(A|B)\cdot\mathbb{P}(B)}{\mathbb{P}(A)}

Boom: Bayes’ theorem. 🖐️🎤💥