The Law of Large Numbers

Flip a coin four times and you might see only heads, but flip it tens of thousands of times and it inevitably settles near a perfect 50-50 split—the ultimate compass of statistics.

Definition The Law of Large Numbers is a foundational principle in statistics stating that as an experiment or observation is repeated over and over, the average of the actual observed results draws closer to the theoretical expected value. While pure luck can distort small samples, repeating the trial enough times reveals the true mathematical rule hidden beneath the randomness.

What Happens When You Flip a Coin 1,000 Times

Imagine pulling a coin from your pocket and flipping it just four times. If all four land on heads by chance, it might look like the probability of heads is 100%. In small sample sizes, random flukes can easily skew the results.

Now, increase those flips to 100, 1,000, or 10,000 times. Even if you started with an unusual streak of heads, the overall ratio steadily marches toward the true mathematical probability of 50%.

Temporary imbalances from early attempts get diluted into the overall average as the number of trials grows. Without any outside intervention, simply gathering more data naturally pulls the results back to where they belong.

Law of Large Numbers: Ratio of heads converges to 50% as trials increase 100% 50% 0% 1 10 100 1,000 % Heads Flips → Expected (50%) Early: Random swings Later: Hits 50%

Why Casinos and Insurers Never Go Broke

A player at a roulette wheel or slot machine might walk away with a huge jackpot on pure luck. For an individual player, gambling feels like a game of chance. But the moment thousands of players place millions of bets, everything turns into cold, predictable math. Casinos hold a tiny mathematical edge of around 1% to 2% per game, and the Law of Large Numbers guarantees that this edge converts into massive, steady profit over time.

Insurance companies operate on the exact same playbook. An insurer cannot predict whether you personally will get into a car crash tomorrow. However, across hundreds of thousands of drivers, they can calculate with pinpoint accuracy what percentage will file a claim this year.

This principle turns uncontrollable individual uncertainty into a predictable, stable risk management model. Modern pillars of society—from big data analytics to census forecasting—rest firmly on this mathematical foundation.

A Closer Look: Watch Out for the Gambler's Fallacy

There is a common trap people fall into when learning about the Law of Large Numbers. If a coin lands on heads 10 times in a row, does that mean tails is 'due' next to balance things out? Absolutely not. A coin has no memory; the odds for the next flip remain exactly 50-50.

The Law of Large Numbers does not compensate for past streaks by forcing opposite outcomes in the future. Instead, the sheer volume of future trials naturally drowns out past anomalies. Even with a 10-heads head start, adding 100,000 more flips turns those 10 early heads into a statistical blip of less than 0.01%.

In mathematics, this is categorized into the 'Weak Law' and the 'Strong Law' of Large Numbers. The Weak Law states that the sample average converges toward the expected value in probability, while the Strong Law proves that as trials approach infinity, the sample average almost surely equals the true expected value.

🤔 Common misconceptions

✕ Myth

According to the Law of Large Numbers, after a long streak of heads, tails is more likely to appear next to balance things out.

✓ Fact

Each independent trial is completely unaffected by past outcomes. The Law of Large Numbers does not 'correct' past imbalances; it simply dilutes them across a massive number of future trials.

🧺 Where you meet it

1 Flipping a coin thousands of times brings the ratio of heads to tails closer and closer to 1:1.
2 Insurance companies accurately calculate annual claim rates across millions of policyholders to set sustainable premium rates.
💡 In one sentence

As the number of trials increases, the average of the results gets closer to the theoretical expected value.