Coin flip
Heads or tails? Flip a single coin or a whole batch — fair, instant, and right in your browser.
Flip a coin online
Tap Flip for a fair, random heads-or-tails result. Need to settle several decisions or run a quick experiment? Set the count higher and flip many coins at once — the tally shows how many landed heads versus tails.
Each flip uses your browser's secure random generator, so it's genuinely 50/50. For other random tools, try the dice roller, the random number generator, or the random name picker.
Is it actually random?
Yes — each flip uses the browser's built-in random number generator to pick heads or tails with an even 50/50 chance and no hidden bias. Flipping many at once is a quick way to see randomness in action: over a handful of flips you'll often see lopsided streaks, but as the count climbs the split trends toward half-and-half — the law of large numbers at work.
Handy for more than bets
A coin flip is a fast, fair tie-breaker for any two-way decision, a simple classroom demo of probability, or a way to make yourself commit when you're stuck between two options — sometimes the flip just clarifies which result you were quietly hoping for. It runs instantly in your browser.
Streaks are normal, not evidence of bias
A run of five heads feels like something has gone wrong, but random sequences are streakier than intuition allows. Flip a fair coin twenty times and there is roughly a 77% chance of seeing at least one run of four or more in a row. Runs are the expected behaviour, not a malfunction.
The related error is the gambler's fallacy — believing that after four heads, tails is "due". A coin has no memory. The chance on the next flip is 50/50 regardless of what came before, because each flip is independent of every other one.
What an even split actually looks like
Over ten flips, an exact five-five split happens only about 24.6% of the time — so roughly three times in four, a fair coin gives you something other than the "expected" result. Getting seven or more heads out of ten happens about 17.2% of the time, which is common enough that it tells you nothing at all about the coin.
What the law of large numbers actually says is that the proportion converges on half as the number of flips grows, not that the counts even out. Over 10,000 flips you would expect the split to sit very close to 50%, while the absolute gap between heads and tails typically grows rather than shrinks. If you want to see this, set the count high and watch the tally.
Where the randomness comes from
Each flip calls crypto.getRandomValues(), the browser's cryptographically secure random number generator, and takes the parity of the result. That is a stronger source than the usual Math.random(), which is fast but not designed to be unpredictable.
It runs entirely on your device — no result is generated on or sent to a server, so nothing about your flips leaves the browser. Because the generator draws on operating-system entropy rather than a seed you control, results are not reproducible: refreshing gives a genuinely new sequence, which is the point for a decision or a giveaway. If you need repeatable pseudo-random output for testing instead, you want a seeded generator, not this.