RNGdle Random Roll Sandbox

A uniform draw gives every valid integer one equal chance out of 1,000,001. This includes zero and one million, which are sometimes accidentally excluded by a range calculation. The sandbox starts from a cryptographic random 32-bit value supplied by the browser. Because 2³² is not an exact multiple of 1,000,001, simply taking a remainder would give some outcomes a slightly larger share. The implementation rejects the small high-end remainder of the 32-bit range and draws again, then maps the accepted value into the number domain. This is rejection sampling.

The important word is independent. The next sandbox number is not corrected to make a previous streak look balanced, and the button does not search for an interesting digit pattern. If the same number appears again, that is allowed. If a run of modest scores appears, the draw does not owe you a rare one. The rule engine examines the number only after selection. EP is an interpretation of the selected integer, never an input to the draw.

Theoretical probabilities can still be useful for understanding a tier. The complete rule enumeration puts 10,000 of 1,000,001 possible integers in the mythic number tier. That ratio is a theoretical fraction of the finite domain, not a forecast about what you will see in a small set of clicks. It also does not say which exact number you will receive. A single specified integer remains one outcome among 1,000,001 each time.

Once a number is selected, the same analyzeNumber function used by the homepage checker evaluates its badge conditions. Some look at arithmetic properties such as powers, primes, factorial values, divisibility, and Fibonacci membership. Others inspect decimal digits: repeated pairs, blocks that mirror one another, ordered runs, or a sum of the digits. A number can match both kinds at once. The sandbox displays all matching badges so the total is inspectable rather than appearing as an unexplained number.

Each badge has a fixed EP value in the captured rule set. The calculation adds independent badges directly. When two or more matching badges share a family, only the highest-valued badge in that family adds its EP; the other matches remain listed but are labeled as not counted. This is especially relevant for numbers that contain nested patterns. A long repeated block often contains shorter repeats as well, and counting all related badges without the family rule would overstate the game’s displayed EP.

Finally, the total EP maps to one of seven number tiers. The thresholds come from the all-number distribution, not from how many people happened to roll a number today. A sandbox result therefore has the same calculated EP and tier as the homepage checker would show for that integer under this captured version. To compare every possible outcome, read the theoretical ranking; to examine a dated set of player results, read the saved day board.

The sandbox is useful when you want to see how often particular patterns appear in your own small experiment or to practice reading badge and family outcomes. A short run can be entertaining, but it is a poor estimator of rare events. The complete domain calculation is the proper source for exact counts. The sandbox neither predicts a future official roll nor improves the chance of receiving high EP. Those two statements remain true whether you click once or many times.

The original game’s daily roll and this browser tool also have different records. An official result may be attached to an account or appear in a public ranking under that game’s rules. A sandbox result is local to your browser. Clearing local history removes the convenience list but does not change the mathematical score for any number. If you want to check a specific result again, enter it in the number checker; the same independent rule function will reproduce its badges and EP.