Wheel Probability Explained: Odds, Repeats and Streaks
The probability of a wheel landing on a slice is that slice’s share of the wheel: 1 ÷ the number of slices when they’re equal, or the slice’s angle ÷ 360° when they’re not. Every spin is independent, so the wheel has no memory. This guide shows how to work out a wheel’s odds, why repeats and streaks happen far more often than people expect, and how to check whether a wheel is fair, with worked examples and real numbers.
The Quick Version
- Equal slices: chance = 1 ÷ number of slices. Ten names gives each one 10%.
- Unequal slices: chance = slice angle ÷ 360°, or the slice’s weight ÷ the total of all weights.
- Several spins: chance of at least once = 1 − (1 − p)n.
- Two things in a row: multiply the chances, because spins are independent.
- Repeats are normal. Picking 10 times from 30 names with repeats allowed gives an 81.5% chance that some name comes up twice.
How to Work Out the Probability of a Spinner
Equal slices
If every slice is the same size, each has the same chance: 1 ÷ the number of slices. A wheel with 4 colours gives each colour 1/4, or 25%. If two of those 4 slices are red, red has 2/4, or 50%.
Unequal slices
Measure the angle of the slice and divide by 360°. A 90° slice is 90/360 = 25%. A 36° slice is 10%. On a digital wheel with weights, divide the entry’s weight by the total weight: weights of 1, 2 and 7 add up to 10, so the chances are 10%, 20% and 70%.
| Wheel | Slice | How to work it out | Chance |
|---|---|---|---|
| 8 equal slices | Any one | 1 ÷ 8 | 12.5% |
| 8 equal slices, 3 are blue | Blue | 3 ÷ 8 | 37.5% |
| A 120° slice | That slice | 120 ÷ 360 | 33.3% |
| Weights 1, 2, 7 | Weight 7 | 7 ÷ 10 | 70% |
Want to set exact odds yourself? Our spin wheel with percentages lets you type *25% after an entry and sizes the slice to match.
The Odds Over Several Spins
A single spin is easy. People usually want to know something about a series of spins: “If I spin 20 times, will the 5% prize come up?”
At least once
The chance of an outcome appearing at least once in n spins is 1 − (1 − p)n, where p is its chance on one spin. A 1-in-6 slice spun 6 times comes up at least once 66.5% of the time, not 100%. Try your own numbers:
Two results in a row, or two wheels
Multiply the chances. Landing on blue (1/4) and then on red on a 3-slice wheel (1/3) is 1/4 × 1/3 = 1/12. The same rule gives the chance of the same name twice in a row on a 10-name wheel: the first spin can be anything, and the second must match it, so it’s 1/10.
Why Repeats and Streaks Happen So Often
Most complaints that a wheel is “rigged” come from repeats and streaks. They feel unlikely, but the maths says they are normal.
| Situation | Chance |
|---|---|
| A name repeats when picking 4 times from 10 names | 49.6% |
| A name repeats when picking 5 times from 30 names | 29.6% |
| A name repeats when picking 10 times from 30 names | 81.5% |
| A run of 4 or more of the same answer in 20 yes/no spins | 76.8% |
| A run of 5 or more of the same answer in 20 yes/no spins | 45.8% |
The repeat figures are the same maths as the famous birthday problem. The streak figures come from our simulation of 200,000 sets of 20 spins. If repeats aren’t what you want, turn on “Remove winner after each spin” so every name comes up once per round, as on the random name picker.
How many spins until every name comes up?
With repeats allowed, it takes far longer than the number of names for every name to appear at least once. On average you need about 15 spins for 6 names, 29 spins for 10 names and 120 spins for 30 names. That’s why teachers who want everyone to get a turn should use no-repeat mode rather than spinning and hoping.
The Gambler’s Fallacy: The Wheel Has No Memory
After a slice hasn’t come up for a while, it can feel “due”. It isn’t. Each spin of a fair wheel is independent, so a 10% slice has a 10% chance on the next spin whether it came up last time or hasn’t appeared in 50 spins. Long gaps and clusters are part of randomness, not signs that the wheel is about to correct itself.
Expected Value of a Prize Wheel
If slices pay out different amounts, the expected value is what one spin is worth on average: multiply each prize by its chance and add them up.
| Prize | Chance | Prize × chance |
|---|---|---|
| $1 | 50% | $0.50 |
| $5 | 30% | $1.50 |
| $20 | 15% | $3.00 |
| $100 | 5% | $5.00 |
| Expected value per spin | $10.00 |
If players pay more than the expected value to spin, the wheel makes money on average; if they pay less, it loses. For prizes in a sales promotion, publish the odds and check the local rules for prize promotions.
Theoretical vs. Experimental Probability
Theoretical probability is what the maths says: a 25% slice should win 25% of the time. Experimental probability is what actually happens when you spin. With a few spins the two can be far apart. With thousands they get close, which is the law of large numbers.
Our wheels have a Fairness tab that runs thousands of spins in seconds. On a wheel set to 5%, 15%, 30% and 50%, 100,000 test spins came out at 4.93%, 15.13%, 30.03% and 49.91%. That makes a good classroom activity: predict, test 10, 100 and 1,000 spins, and compare. See our classroom wheels for more.
Are Real Spinning Wheels Fair?
A physical wheel is only as fair as its build. Slices that aren’t quite equal, a tilted stand, a sticky pointer or friction that stops the wheel in the same zone can all shift the odds. A strong, consistent spin from the same starting point can also make results more predictable on a cheap wheel.
- Check that slices really are equal, or measure their angles.
- Spin hard enough for several full turns, and start from a random position.
- Keep a tally of 100 or more spins and compare it with the expected counts.
A digital wheel avoids these problems if it picks the winner with a proper random number generator. Ours draws the result with your browser’s secure generator before the animation starts. Read how our draws work and how to tell if a wheel is rigged.
Frequently Asked Questions
How do you solve spinner probability?
Count the slices that give the result you want and divide by the total number of equal slices. For unequal slices, divide the slice’s angle by 360°. For several spins, multiply chances for “and”, and use 1 − (1 − p)n for “at least once”.
Is spin the wheel actually random?
A digital wheel is random if it uses a good random number generator to pick the winner. Ours uses a cryptographic generator, and you can test it with thousands of spins in the Fairness tab. A physical wheel depends on how well it’s made and spun.
Why did the same name come up twice?
Because each spin is independent. On a 10-name wheel, the chance of the same name twice in a row is 1 in 10, and over several spins some repeat is very likely. Use no-repeat mode if every name should come up once.
How can I spin a wheel and win a prize?
On a fair wheel you can’t influence the result. Your chance is the share of the wheel taken by the prize slices. On a game show or fairground wheel, the rules and the wheel’s build set the odds, so ask what they are.
Does a bigger slice guarantee a win?
No. A 70% slice still loses 30% of the time, and can lose several spins in a row.
Do weights and percentages mean the same thing?
A percentage is a share of the whole wheel. A weight is a number of chances compared with the others. Weights of 1, 2 and 7 give 10%, 20% and 70%; doubling every weight changes nothing.
Sources
- NCTM Illuminations: Adjustable Spinner (theoretical and experimental probability)
- Figures for repeats use the birthday-problem formula; streak figures are from our simulation of 200,000 sets of 20 fair yes/no spins; “every name” figures are the coupon collector’s expected value, n × (1 + 1/2 + … + 1/n).