Probability — measuring what hasn't happened

There is a reason probability feels uniquely hard. Everywhere else in mathematics, intuition helps; here, intuition betrays you to your face. So we will not fight your intuition — we will let you catch it in the act, and then show you a method that needs no intuition at all.

First handle — memory

Nine heads in a row. And the tenth?

Here is a coin that just came up heads nine times straight. If a voice inside says "tails is due now" — that is normal. Most people hear it. Now toss the tenth flip yourself. As many times as you like.

H H H H H H H H H ·

?

The more you stack, the closer to half-and-half it settles. Tails was never "due" — the coin knows nothing about the previous nine flips. A thing without memory cannot have turns. This illusion — the gambler's fallacy — is not something to be embarrassed about; it is human intuition's factory setting. Which is exactly why probability had to be invented as calculation: here, feelings cannot be trusted.

Second handle — the denominator

Counting instead of guessing

Roll two dice and look at the sum. Sum 7 and sum 12 — "both just one number, so shouldn't their chances be similar?" Let's check. The whole of probability is one fraction: (ways it happens) ÷ (all the ways). The grid below is that denominator — all thirty-six ways, in plain sight.

Six roads lead to a sum of 7; only one leads to 12 — the same "single number", but a different number of roads. The moment the denominator 36 is visible, intuition is no longer needed. You just count. When people get lost in probability, it is almost never the numerator — it is losing sight of how many squares "all the ways" contains.

This calculation was born from a gambling dispute

Until chance was calculated

Dice are thousands of years old. Calculating chance is not yet four hundred.

The letters — Pascal and Fermat

how to split the stakes

In 1654 two mathematicians exchanged letters about how to divide the stakes of a gambling game interrupted midway. How to price a round that has not happened yet — probability theory was born in those letters.

The scale — from 0 to 1

a ruler for possibility

Impossible is 0, certain is 1. Once every possibility in between got a mark on that ruler, words that used to blur into "maybe" and "who knows" became comparable numbers. The forecast's 70% is one notch on this ruler.

Today — AI

machines speak in probabilities

The AI that helped build this page chooses every word by probability. Insurance premiums, drug approvals, your navigation's arrival time — modern judgment mostly stands on this one fraction. Reading probability has become reading the world.

A calculation born from gambling debts became, within four centuries, civilisation's standard of judgment. It was needed where intuition betrays us — and being needed there made it the most useful mathematics of all.

Words around chance

Where this calculation goes beyond coins and dice.

Browse the full glossary →

Where this sits on the map

There are no exercises here. If you caught the tenth coin knowing nothing about the first nine, and counted the thirty-six squares with your own eyes, this page has done its job.

If the coin remembers nothing, a reverse question appears — when I come to know something, do the chances change? What is the probability that a person whose first test came back positive actually has the disease? That answer is conditional probability, and at the end of that road waits "Why AI answers in probabilities". The faint nodes on the map — that is where we go next.

See this node on the map →