Flip Multiple Coins

How the multi-coin flipper decides each coin

When you press Flip them, each coin is decided on its own with a fresh 50% chance of heads. The coins do not influence each other, so tossing 100 coins at once gives exactly the same odds as flipping one coin 100 times. The tool then counts heads and tails and works out how likely that exact count was, using the same binomial formula as the probability calculator.

The expected number of heads is half the number of coins. The typical spread around that center grows with the square root of the number of coins: for 100 coins, the standard deviation is 5 heads. So results between 45 and 55 heads are routine and results between 40 and 60 cover about 96.5% of all tosses.

A worked example with four coins

Four coins can land in 16 equally likely patterns, because each coin doubles the possibilities: 2 x 2 x 2 x 2 = 16. Count how many patterns give each number of heads. One pattern gives no heads, four give one head, six give two heads, four give three and one gives four. Those counts, 1, 4, 6, 4, 1, are the fifth row of Pascal's triangle. So the chance of exactly two heads is 6 in 16, or 37.5%. The chance of all four landing heads is 1 in 16, or 6.25%.

The same counting works for any number of coins, but the rows grow fast. With 10 coins there are 1,024 patterns and 252 of them have exactly five heads, which is where the 24.6% figure on this page comes from.

Why the "chance of this" number looks small

The more coins you toss, the less likely any single exact count becomes, even the most likely one. With 100 coins the chance of exactly 50 heads is only about 7.96%. With 1,000 coins the chance of exactly 500 heads falls to about 2.5%. That does not mean the result is unusual. It means there are many possible counts sharing the probability. To judge whether a result is surprising, look at how far it is from half, measured in standard deviations, rather than at the exact-count percentage.

Famous coin-tossing experiments

People have tested coin tossing by hand for centuries. The statistician Karl Pearson recorded 24,000 tosses and got 12,012 heads. The mathematician John Kerrich, held in an internment camp in Denmark during the Second World War, tossed a coin 10,000 times and recorded 5,067 heads. Both results sit comfortably within the normal spread. The multi-coin flipper lets you repeat a small version of their experiment in seconds: flip 100 coins ten times, add up the heads and see how close to 500 you land.

Ways to use it

Teachers use multiple coins to build probability distributions. Have each student flip 10 coins, record the number of heads and plot the class results. The chart will pile up around five and thin out toward zero and ten. Game players can use it as a stand-in for dice: three coins give eight equally likely patterns. It also works for quick random choices across a group, where each person is assigned a coin and heads means in. Because the grid shows every coin in order, you can read the result coin by coin rather than only as a total.

Common mistakes

The usual mistake is expecting an even split every time. With an even number of coins, the exact half split is the single most likely result, but it still loses to "anything else" most of the time. Another is comparing percentages across different sizes. Getting 60% heads is common with 10 coins and rare with 100. A third is treating a run of lopsided results as a sign the tool is broken. Each toss is independent and the next one does not correct the last. Over many tosses the average settles toward half because the early results are diluted, not because later coins push back.

Frequently asked questions