Flip a Coin

The Mathematics of a Coin Flip

A fair coin has a 50/50 probability of landing heads or tails on any given flip. But "fair" is doing a lot of work in that sentence. Real coins are not perfectly fair — a 2007 Stanford study by Persi Diaconis found that a coin is actually about 51% likely to land on the same face it started on, due to the physics of the flip. The bias is too small to exploit practically, but it means the coin flip is not the perfectly random event most people assume.

Over a large number of flips, results converge toward 50/50 with remarkable precision. This is the Law of Large Numbers in action — not a guarantee that results will be exactly even, but a mathematical proof that the ratio of heads to tails approaches 0.5 as flips increase. Flip a coin 10 times and getting 7 heads is unremarkable. Flip it 10,000 times and getting 7,000 heads would be astronomically unlikely — essentially impossible if the coin is fair.

The Gambler's Fallacy

After five heads in a row, most people feel that tails is "due." This is the gambler's fallacy — the belief that past random outcomes influence future ones. The coin has no memory. The probability of heads on the sixth flip is still exactly 50%, regardless of what happened on flips one through five. Five heads in a row is uncommon (3.125% probability) but once it has happened, it tells you nothing about flip six.

This fallacy costs real money in casinos, sports betting, and financial markets. Traders who believe a stock is "due for a bounce" after falling for five consecutive days are applying the same flawed logic. Each day's price movement is influenced by new information and market conditions, not by a cosmic need to balance prior movements.

Using Coin Flips for Decisions

Here is the best decision-making trick involving a coin: flip it, but pay attention to your emotional reaction when you see the result. If you feel relief, the coin made the right call. If you feel disappointment, you wanted the other option — and now you know your true preference. The coin's actual result does not matter. It is a tool for bypassing the paralysis of overthinking by revealing what your gut already decided.

Is this truly random?

This uses your browser's cryptographic random number generator (crypto.getRandomValues), which produces results indistinguishable from true randomness for practical purposes. It is as fair as a digital coin flip can possibly be.

What are the odds of getting heads 10 times in a row?

0.5^10 = 0.0977%, or roughly 1 in 1,024. Unlikely on any given set of 10 flips, but if you flip a coin 10,000 times, you will see runs of 10 identical results roughly 5 times. Rare events happen regularly when you have enough trials.