Skip to content
ToolsNow
Guides All tools
Random

Coin flip — heads or tails, a true 50/50

Flip a coin — or a hundred at once — with a truly 50/50 cryptographic flip, a running tally, and the heads percentage as it converges.

Runs in your browser — your input is not uploaded to ToolsNow.

Flip to get heads or tails.

The result is generated locally in your browser.

Each flip is an independent 50/50 draw from the browser's cryptographic generator. The running tally is there to watch the law of large numbers do its thing: short runs of several heads in a row are completely normal, and the percentage only settles near 50% over many flips. Past flips never influence the next one — the gambler's fallacy is believing otherwise.

How to use

  1. Flip. One coin for a decision, many for an experiment.
  2. Read the result. Heads or tails, with each coin shown.
  3. Watch the tally. The running percentage drifts toward 50% as flips accumulate.

Questions

Is this really 50/50?

Exactly 50/50, from the browser’s cryptographic generator — which is more than can be said for a physical coin: studies of real coin flips find the side facing up at launch lands up about 51% of the time, because of precession in the toss.

I got five heads in a row. Is it broken?

No — runs are what genuine randomness looks like. Five heads in a row has a 1-in-32 chance on any five flips, so it happens regularly. Each flip is independent: after five heads, the next flip is still exactly 50/50. Feeling that tails is now “due” is the gambler’s fallacy.

Why does the percentage settle near 50% but the counts drift apart?

The law of large numbers guarantees the ratio converges, not the difference. After 10,000 flips a gap of 100 between heads and tails is normal — that is a 1% difference in share — and the absolute gap typically grows with the square root of the number of flips even as the percentage tightens.

Can I use this to decide something important?

A coin flip is a time-honoured tie-breaker because it is incorruptible and instant. There is also a real psychological trick in it: if the result disappoints you, you have just learned what you actually wanted — at the cost of one flip.

Examples and checks

Use a single flip for an informal choice or batches to explore probability. The running totals describe the flips already made; they cannot predict the next result.

Interpret a short streak

Input or situation
Five flips all land heads.
Result
That particular sequence has probability 1 in 32.

Multiply five independent one-half probabilities: (1/2)⁵. The next flip still has a one-half probability of heads.

Compare a batch

Input or situation
100 independent flips
Result
50 heads is the expected count, not a guaranteed count.

Repeated batches vary. A result above or below 50 is expected and does not by itself show that the generator is biased.

Before you use the result

Published by ToolsNow. How tools and sources are checked · Report a problem

Related tools