Coin Dynamics

Bernoulli trials at scale, from a single flip to one million.

Flip a virtual coin up to large sample sizes and observe how short-term streaks gradually converge toward the long-run 50/50 probability, a live demonstration of the law of large numbers.

Options
Coin landing on edge(probability 0.012%, or 1 in ~8,333 flips). Documented in real physics research.
?
Mass Simulation

Runs entirely in your browser, no server involved. Up to 1,000,000 flips processed in real time.

Statistics
0 rolls this session

No flips this session.

Frequently Asked Questions

Is a coin flip truly 50/50? ▾

A perfectly fair coin flip is 50/50. However, research shows physical coins have a slight bias toward the side facing up when tossed, due to the dynamics of the flip. The effect is small (~51% vs 49%) and only detectable over thousands of flips.

What does the chi-square test show? ▾

The chi-square test measures whether the observed distribution of heads and tails differs significantly from the expected 50/50. A p-value above 0.05 means the results are consistent with a fair coin. A p-value below 0.05 suggests the distribution is unusually skewed.

Can you get 10 heads in a row with a fair coin? ▾

Yes. The probability of 10 consecutive heads is (0.5)^10 = 1/1024 ≈ 0.098%. Rare, but not remarkable. Over 10,000 flips you would expect this to happen roughly 10 times. Streaks feel meaningful but are a normal feature of random sequences.

What is the Law of Large Numbers? ▾

The Law of Large Numbers states that as the number of trials increases, the observed frequency converges toward the theoretical probability. Flip 10 coins and you might see 7 heads. Flip 1,000,000 and the proportion will be very close to 50%. Use the 1M flip option to see this in action.

Related Reading

More Tools