Calculate coin flip probabilities, simulate flips, and explore binomial distribution for heads/tails outcomes. Streaks, runs, and fairness testing included.
The Coin Flip Probability Calculator computes exact probabilities for coin-flip experiments with a customizable coin bias. It is built around the binomial model, so it can answer questions like "What is the chance of exactly k heads?" as well as cumulative questions such as "What is the chance of at least k heads?" It is a simple way to turn repeated coin flips into a real probability model instead of a guess.
Enter the number of flips, the desired outcome count, and a bias value to view exact probabilities, a simulated sequence, and streak analysis. That makes it useful for classroom probability, informal games, and quick checks of random-looking sequences.
The fairness test checks whether the observed results fit a fair coin using a chi-squared approach. You can also explore how likely streaks are, which is helpful because humans tend to underestimate runs of the same outcome. It is a handy way to separate random variation from a real bias.
Use this calculator for games, teaching, fairness checks, and binomial probability experiments. It combines exact binomial results with simulations, streak estimates, and a fairness test, which makes it useful both for classroom examples and for checking whether a result is actually unusual. It is especially helpful when you want to compare intuition about streaks with the actual binomial odds.
P(X = k) = C(n,k) × p^k × (1-p)^(n-k). Where: C(n,k) = n! / (k! × (n-k)!). Expected Heads = n × p. Variance = n × p × (1-p). Streak of m heads probability ≈ 1 - (1 - p^m)^(n-m+1) (approximation).
Result: P(X = 7) = 11.72%, P(X ≥ 7) = 17.19%
With 10 fair coin flips, P(exactly 7 heads) = C(10,7) × 0.5^10 = 120/1024 = 11.72%. P(at least 7 heads) = P(7) + P(8) + P(9) + P(10) = 17.19%.
The binomial coefficient C(n,k) counts the number of ways to arrange k heads among n flips. For n = 10, k = 5: C(10,5) = 252 arrangements, each with probability 0.5^10 = 0.00098. So P(exactly 5 heads) = 252 × 0.00098 = 24.6%.
The distribution is symmetric for fair coins and skews toward the higher-probability outcome for biased coins. The central limit theorem says that for large n, the distribution of heads approaches a normal distribution.
Humans consistently underestimate streak lengths in random sequences. In 200 fair coin flips, the expected longest streak of either outcome is about 7-8 flips. Research shows people expect random sequences to alternate more than they actually do.
This misperception drives the "hot hand" debate in sports: are shooting streaks real or statistical artifacts? The answer: some streaks exceed random expectation, but many are within normal variance.
Given a biased coin with P(H) = p and P(T) = 1-p: flip twice. If HT, output heads. If TH, output tails. If HH or TT, discard and reflip. P(HT) = p(1-p) = P(TH), so the outputs are perfectly fair regardless of p. Expected flips per output = 2/(2p(1-p)).
With a fair coin, the probability is 0.5^10, or about 0.0977%. That is rare for one specific run of 10 flips, but streaks become much less surprising when you look at a long sequence of flips instead of a single segment. In 1,000 flips, a streak of that length is no longer unusual.
A mathematically fair coin is 50/50, but physical coins can show small biases depending on how they are tossed and caught. Real-world research has found slight tendencies in some conditions, although the effect is usually small enough that it does not matter for ordinary classroom or game use. The calculator lets you model either a fair or biased coin.
You generally need a lot more flips than people expect, because small deviations can happen by chance. A rough test of fairness needs enough data to separate random variation from a real bias, and the chi-squared test is one common way to do that. More flips give you a more reliable answer, especially when the suspected bias is small.
For fair coin flips, the longest streak tends to grow slowly with n, which is why streaks feel more surprising than they really are. In short sequences you might see runs of 3 or 4, while much longer runs become plausible as the number of flips increases. The calculator helps you estimate those run lengths instead of guessing.
Each flip is a Bernoulli trial, meaning it has two possible outcomes and a fixed success probability p. The total number of heads in n flips follows a Binomial(n, p) distribution. That model is the foundation for exact probability calculations in coin-flip problems and many other yes-or-no experiments.
It is the false belief that past flips make the next flip more likely to come up the opposite way. A fair coin does not remember earlier results, so a streak of heads does not make tails "due." Random sequences naturally contain runs, and that is part of what this calculator helps make visible.