Free Dice, with the formula on the table
Free Dice on the Stake-style formula: pick a win chance from 0.01% to 98%, roll over or under, and watch 99 ÷ chance become the multiplier. Play money, provably fair, no account.
- Win chance
- 50.00%
- Multiplier
- 1.9800×
—
Roll to start
See the maths
One float from the HMAC stream: roll = floor(float × 10001) / 100, giving 0.00–100.00 in equal steps.
Your win chance is the share of that range you cover. The multiplier is 99 / chance% — the 99 (not 100) is the whole house edge: 1%, at every setting.
Play money only. Nothing to buy, nothing to win, nothing leaves your browser. Reset the balance whenever you like.
Chance to multiplier, and the streaks that come with it
Slide the win chance and two things move together: the multiplier falls as the chance rises, and the losing streaks get shorter. The expected return never moves. chance × (99 ÷ chance) = 99% whatever you pick.
| Win chance | Multiplier | P(lose a roll) | P(10 losses in a row) | Ten losses, one in | Expected rolls per win | Return |
|---|---|---|---|---|---|---|
| 2% | 49.5000× | 98.0% | 81.71% | 1 in 1.2 | 50.0 | 99% |
| 5% | 19.8000× | 95.0% | 59.87% | 1 in 1.7 | 20.0 | 99% |
| 10% | 9.9000× | 90.0% | 34.87% | 1 in 2.9 | 10.0 | 99% |
| 25% | 3.9600× | 75.0% | 5.63% | 1 in 18 | 4.0 | 99% |
| 49.5% | 2.0000× | 50.5% | 0.11% | 1 in 927 | 2.0 | 99% |
| 66% | 1.5000× | 34.0% | 0.002064% | 1 in 48,441 | 1.5 | 99% |
| 75% | 1.3200× | 25.0% | 0.000095% | 1 in 1.0 × 10^6 | 1.3 | 99% |
| 90% | 1.1000× | 10.0% | 1.0e-8% | 1 in 1.0 × 10^10 | 1.1 | 99% |
| 95% | 1.0421× | 5.0% | 9.8e-12% | 1 in 1.0 × 10^13 | 1.1 | 99% |
| 98% | 1.0102× | 2.0% | 1.0e-15% | 1 in 9.8 × 10^16 | 1.0 | 99% |
How Dice works
One float, one roll. The round computes HMAC-SHA256(server seed, client seed:nonce:0), reads the first four bytes as a float between 0 and 1, and the roll is floor(float × 10001) / 100: a number from 0.00 to 100.00 with 10,001 equally likely values.
Your chance is the share of the range you cover. Roll over 50.50 covers 4,950 of the 10,001 values, which is 49.5%. Roll under 2.00 covers 200 values, which is 2%. The game shows the chance and the multiplier before you roll.
The 99 is the edge. A fair game would pay 100 ÷ chance. This one pays 99 ÷ chance, so one percent of every stake is expected to stay with the house, at every setting, on every roll. The Dice strategy guide works through what that does to martingale and its cousins.
Questions
- Why does 49.5% pay exactly 2× and not 50%?
- The multiplier is 99 divided by the win chance in percent, so a 49.5% chance pays 2.0000×. A fair coin would need 50% for 2×. That missing half a percentage point, spread over every roll, is the whole 1% house edge: the expected return at any chance is chance × 99 / chance = 99% of the stake.
- How likely is a long losing streak?
- At 49.5%, ten losses in a row happen 0.11% of the time you start counting (1 in 927), and over a session of hundreds of rolls a run like that is common, not rare. At 2% win chance a losing streak is the normal state: the expected wait for one win is 50 rolls and ten straight losses happen 81.7% of the time.
- Is the free game rigged to pay more than the real one?
- No. The roll is floor(float × 10001) / 100 from an HMAC-SHA256 float, the same derivation Stake documents, and the multiplier is the same 99 / chance. A practice game that paid better than the real one would teach the wrong lesson. Your session stats page shows the return you actually realised against the 99% expectation.
Keep going
Last reviewed 2026-09-27