A fair coin flip has a 0.5 probability of landing on either heads or tails, and a streak of ten identical outcomes in a row has a probability of exactly 1 in 1,024. Those two numbers — the constant single-flip probability and the shrinking streak probability — are the foundation of every coin flip cheat sheet, and both follow directly from the fact that a fair coin is a Bernoulli trial with two equally likely outcomes.
This cheat sheet pulls together the core facts you need when you want to flip a coin online, settle a 50/50 decision, or sanity-check a long run of results. It covers the probability math behind a single toss, the streak frequencies you should expect across hundreds of flips, and the exact clicks to run a fair virtual coin in your browser. Bookmark it for any moment a binary choice shows up and you want a quick reference instead of a long explanation.
A virtual coin flip is useful whenever a real coin is not available, when you want the result to be provably unbiased, or when you want to record a session for later review. The reference material below condenses that information into the formulas, frequency tables, and step-by-step instructions you can return to any time you need a one-screen summary.
Coin Flip Probability at a Glance
Every single flip of a fair coin is a Bernoulli trial with probability 0.5 for heads and 0.5 for tails. That constant does not change between flips, so the probability of getting heads on the next toss is always 0.5, no matter what came before. The table below lists the most useful streak probabilities for a fair coin — values you can compute by multiplying 0.5 by itself for each consecutive result you want.
| Streak length (same side) | Calculation | Probability | Approx. odds |
|---|---|---|---|
| 2 in a row | 0.5 × 0.5 | 1 / 2 | 1 in 2 |
| 3 in a row | 0.5³ | 1 / 4 | 1 in 4 |
| 4 in a row | 0.5⁴ | 1 / 16 | 1 in 16 |
| 5 in a row | 0.5⁵ | 1 / 32 | 1 in 32 |
| 6 in a row | 0.5⁶ | 1 / 64 | 1 in 64 |
| 7 in a row | 0.5⁷ | 1 / 128 | 1 in 128 |
| 8 in a row | 0.5⁸ | 1 / 256 | 1 in 256 |
| 9 in a row | 0.5⁹ | 1 / 512 | 1 in 512 |
| 10 in a row | 0.5¹⁰ | 1 / 1,024 | 1 in 1,024 |
Each row uses the same formula: raise 0.5 to the power of the streak length. Because the exponent grows by one each step, the probabilities shrink by half every time you add another identical result. That doubling is why a streak of five feels surprising while a streak of ten feels almost impossible — even though both are exactly what the math predicts for a coin with no bias.
Worked example for k = 5
If you want to confirm any row by hand, plug the streak length into the formula (0.5)^k. For k = 5: 0.5 × 0.5 × 0.5 × 0.5 × 0.5 = 0.03125, which is 1 / 32, or about 3.125%. Over 1,000 flips you would therefore expect roughly 31 streaks of five identical results just by the math — not because the coin is biased.
How to Flip a Coin Online in 3 Steps
The fastest path from a question to a fair heads-or-tails result is the Coin Flip tool. Everything runs locally in your browser, no account is required, and the result is generated from a cryptographically secure random source instead of an ordinary Math.random call. Follow the three steps below and the session stats will give you a running record of every toss.
- Click the Flip button to toss the coin — the result shows as Heads or Tails.
- Watch the live session stats update: heads count, tails count, total flips, and your current streak.
- Use Flip 10x to toss ten coins at once, or Reset to clear your session and start fresh.
Each click of Flip counts as a single Bernoulli trial in your session. The heads and tails tallies update on every toss, the total reflects how many times you have clicked, and the streak counter increments or resets based on whether the latest result matched the one before it. Flip 10x adds ten independent trials at once, which is a quick way to grow your session totals when you want to compare observed frequencies against the expected 50/50 split.
Streaks, Independence, and the Law of Large Numbers
The single most common reaction to a long streak is suspicion: "the coin must be broken." In a fair sequence, that reaction is wrong more often than intuition expects. Each flip is independent, which means the outcome of the next toss is determined only by the random source — not by what came before. After four heads in a row, the probability of heads on the next flip is still exactly 0.5.
This independence is the engine behind streaks. The probability of getting five heads in a row is the product of five independent events, each with probability 0.5, which is 0.5⁵ = 1 / 32. Once you flip often enough, those rare events stop feeling rare: across 1,000 flips you will see roughly 31 streaks of five identical results, and across 100,000 flips you should expect close to 3,125 streaks of five. None of them is evidence of bias.
The Law of Large Numbers is the second piece. Any individual run can look lopsided — you can flip ten heads to start a session, or watch tails dominate for the first twenty tosses. As the total number of flips grows, the running ratio of heads to tails converges toward 50/50. The session stats on the Coin Flip tool exist precisely to make that convergence visible: heads and tails inch toward parity even when the early sequence looks dramatic.
When the source of randomness is cryptographically secure — the Coin Flip tool draws from the CSPRNG via crypto.getRandomValues — the fairness holds up under statistical testing. A chi-square goodness-of-fit test over 100,000 simulated flips stays comfortably within statistical tolerance of the expected 50/50 split, so the visible streaks are genuinely what fair randomness looks like rather than a software bug.
When a Virtual Coin Flip Is the Right Tool
A coin flip is a clean solution whenever the decision is a true binary and the stakes are low enough that either outcome is acceptable. The scenarios below are the ones where a virtual coin adds the most value over a real coin, a dice roll, or a manual choice.
| Scenario | Why a coin flip fits | What to watch for |
|---|---|---|
| Quick either-or decisions | Removes deliberation and personal bias | Make sure both options are genuinely acceptable |
| Tie-breaking between two options | Provably fair 50/50 outcome | Agree on which side maps to which option first |
| Sports kickoff or referee calls | Cryptographically secure randomness, auditable | Use a single fresh click per official call |
| Classroom probability demos | Live stats make the Law of Large Numbers visible | Run long enough — at least 100 flips — for the convergence to show |
| Decision logging for later review | Session totals and streak counter are visible | Click Reset between unrelated decisions to keep stats clean |
For situations that are not strictly 50/50 — picking among three or more options, weighting some outcomes more heavily, or drawing an integer from a larger range — a Random Number Generator or a Random Name Picker wheel is a better fit. The cheat sheet rule of thumb is simple: if the answer is exactly "Heads or Tails" with no further preference, use a coin; if there are weighted or multi-option choices, reach for a tool that supports those directly.
Quick Reference: Coin Flip vs. Number Generator vs. Yes/No
Three generators in this category all produce quick random answers, but each one fits a different shape of question. The table below summarizes what changes when you swap one for the other.
| Tool | Best for | Output shape | Randomness source |
|---|---|---|---|
| Coin Flip | Strict 50/50 binary decisions | Heads or Tails | Crypto CSPRNG (crypto.getRandomValues) |
| Random Number Generator | Integers from an inclusive range | One integer per draw | Crypto CSPRNG (crypto.getRandomValues) |
| Yes or No Generator | Casual 50/50 prompts with a question label | Yes or No plus history | Browser Web Crypto (local) |
Keep this cheat sheet open in a tab whenever a 50/50 decision comes up. Click Flip for the strict binary, run Flip 10x when you want a quick sample of independent outcomes, and let the session stats prove that the long streaks you saw along the way were always going to happen — even when the coin felt, in the moment, suspiciously one-sided.