Roobet Crash fairness check
Across 113,186 recorded rounds, Roobet Crash is statistically consistent with its published formula and return.
All recorded rounds, judged together with every other check on the monitor.
| Check | Measured (95% range) | Stated | Rounds | Status |
|---|---|---|---|---|
| Shape of the result curvevs the operator's own formula | p 0.90 | Formula curve | 113,186 | Consistent |
| Return at 2x implied by every roundMaximum likelihood under the formula; uses the share of rounds that end at 1.00x | 95.54% (95.41 to 95.67) | 95.52% | 113,186 | Consistent |
| Cash out at 1.5x | 95.29% (94.87 to 95.71) | 95.68% | 113,186 | Consistent |
| Cash out at 2x | 95.18% (94.60 to 95.76) | 95.52% | 113,186 | Consistent |
| Cash out at 5x | 94.63% (93.50 to 95.78) | 95.23% | 113,186 | Too little data |
| Cash out at 10x | 94.49% (92.80 to 96.21) | 95.14% | 113,186 | Too little data |
How we calculate this
Each round ends at one multiplier. For a cash-out at a target, the return is the target times the share of rounds that reached it. We compare that share with what the operator's own formula (Roobet verifier: 1 in 25 rounds at 1.00x, else floor((100e - h) / (e - h)) / 100) predicts, give a 95% range for every figure, and mark targets with too few rounds as insufficient instead of showing a number we can't stand behind. Targets from 20x to 100x appear by themselves once enough rounds have reached them. The implied return uses every round at once: under the formula, the share of rounds that end at 1.00x fixes the return at every target.
Every check gets a two-sided p-value. Because 177 checks run on this monitor at once, the p-values are corrected with the Holm method before a result counts as not consistent (corrected p below 0.01). A result with an uncorrected p below 0.01 is shown as borderline. These thresholds are set in advance and are the same for every game. Full method.
Formula: Roobet verifier: 1 in 25 rounds at 1.00x, else floor((100e - h) / (e - h)) / 100.