What is the fairest way to divide a pot of money when a game of chance is interrupted halfway through?
Imagine you and a friend have each bet $50 on a coin-flipping game. The goal is to be the first to reach 10 points. The score is currently 8 to 6 in your favor. Suddenly, an emergency forces your friend to leave. You don’t want to return their $50 because you are winning, but they won’t agree to hand over the full $100 because they still have a chance to make a comeback.
This dilemma, known as the “problem of points,” stumped the greatest minds in mathematics for over 150 years. The quest to solve it eventually led to the birth of probability theory, a field that now governs everything from the stock market to insurance premiums.
The Failed Solutions: Proportion and Progress
Before the mid-17th century, mathematicians attempted to solve this problem using logic that, while intuitive, was mathematically flawed.
- The Proportional Approach (1494): Luca Pacioli suggested that players should split the pot based on their current score. In our 8–6 example, you would take 8/14ths of the pot ($57.14) and your friend would take 6/14ths ($42.86). However, this fails in extreme cases: if the game is interrupted after just one flip, the winner would take the entire pot, even though the game is far from decided.
- The Progress Approach: Niccolò Fontana “Tartaglia” attempted to solve this by looking at how close a player was to the finish line. He argued that a player’s share should be based on their progress relative to the total points needed. While more equitable than Pacioli’s method, it still failed to account for the actual mathematical likelihood of winning, often resulting in payouts that didn’t reflect the true odds.
The Breakthrough: Pascal and Fermat
The deadlock was broken in the 1650s when a French socialite asked the mathematician Blaise Pascal to solve the problem. Pascal turned to his colleague Pierre de Fermat, and their correspondence changed mathematics forever.
They realized that a “fair” split shouldn’t be based on the score as it stands, but on the possible futures of the game. They arrived at the same conclusion using two different, brilliant methods.
Fermat’s Method: Exhausting All Futures
Fermat proposed looking at every possible way the game could continue. If there are five flips left to decide the game, he would list every single possible sequence of heads and tails. He then counted how many of those sequences resulted in a win for Player A versus Player B.
In our 8–6 scenario, Fermat would calculate that there are 32 possible outcomes for the remaining flips. He found that Player A wins in 26 of those scenarios. Therefore, Player A is entitled to 81.25% of the pot ($81.25).
Pascal’s Method: The Power of Expected Value
While Fermat’s method was accurate, it was impractical for long games. If 20 flips remained, you would have to calculate over a million different futures.
Pascal solved this by working backward using a concept we now call “expected value.” He started with the simplest possible scenario: if the score is tied (9–9), the pot is split 50/50. He then worked one step back: if the score is 9–8, there is a 50% chance the leader wins immediately and a 50% chance they tie. By averaging these possibilities, he could calculate the value of any score, step by step, without having to list every single future.
Why This Matters Today
The convergence of Pascal and Fermat’s methods proved that probability isn’t just about what has happened, but about the weighted average of what could happen.
This shift from looking at the past (the current score) to calculating the weighted possibilities of the future is the foundation of modern risk assessment.
Today, this logic is the engine behind much of our modern world. When an insurance company calculates your premium, or a hedge fund manages a portfolio, they are using the descendants of Pascal and Fermat’s logic to weigh potential losses against potential gains.
Conclusion: By attempting to settle a simple gambling dispute, mathematicians moved beyond mere arithmetic to discover the mathematics of uncertainty, providing the tools necessary to navigate a world driven by risk.
