Wheel Randomness Test
Is the wheel rigged? Don't take our word for it. This test spins a virtual wheel 1,000, 10,000 or 100,000 times right in your browser, using exactly the same code that picks winners on every Wheel Spin Hub wheel, then charts how often each slice came up next to how often it should have.
A fair wheel doesn't give every slice the exact same count. Real randomness wobbles. With 8 slices and 10,000 spins each slice should land about 1,250 times, and gaps of a few percent either way are normal. What matters is whether the wobble is the size chance predicts, and that is what the chi-square test measures. A p-value above 0.01 means the spread looks like honest luck.
The test also checks the animation. For every virtual spin it plans the full rotation, works out which slice ends up under the pointer and compares that with the slice that was picked. "Landed on the pick" should read 100%, which shows the wheel you watch always stops where the draw says.
Turn on weighted slices to check custom odds too. Slices then get 1×, 2× and 3× weight in turn, and the expected marks move to match, the same way weighted entries and prize wheels work on the main wheel.
How to use the Wheel Randomness Test
- Choose how many slices the test wheel has.
- Pick 1,000, 10,000 or 100,000 spins.
- Tick weighted slices if you want to test uneven odds.
- Press "Run the test" and compare each bar with its expected mark.
Tips
- More spins means bars closer to the expected marks: try 100,000 to see the gaps shrink.
- Share this page when someone says the wheel is rigged in a giveaway or classroom.
- Every run uses fresh random numbers, so results differ a little each time.
Frequently asked questions
Is the Wheel Spin Hub wheel rigged?
No. Every spin picks a winner with your browser's random number generator, weighted only by the odds you can see on the wheel. Nothing is stored on a server and no result is decided before you spin. This test lets you check it yourself.
Why aren't all the bars exactly the same height?
Because real randomness varies. A fair coin flipped 100 times rarely gives exactly 50 heads. The chi-square score and p-value tell you whether the differences are the normal size for chance.
What does the p-value mean?
It's the chance a perfectly fair wheel would spread out at least this unevenly. Above 0.01 is normal. A fair wheel still drops below 0.01 about once in 100 runs, so run it again before drawing conclusions.
Does the test use the same code as the real wheel?
Yes. It calls the same functions that pick the winner and plan the spin on every wheel on the site, and it also checks that the slice under the pointer matches the pick.
Can a wheel be made to favour one entry?
Only openly. Turn on weighted entries in advanced settings and the favoured entry gets a visibly bigger slice. There is no hidden setting that changes the odds.