Random Number Generator

Where these numbers come from

This uses crypto.getRandomValues(), the browser's cryptographically secure random source, which draws entropy from the operating system. That is a genuinely different thing from Math.random(), which is a fast pseudo-random generator with internal state that can, in principle, be inferred from enough output. For picking a raffle winner either is fine; for anything where someone has a reason to predict the result, only the first will do.

The generation happens entirely on your device. No request is made, no draw is logged, and the page works offline.

The modulo bias, and why it is avoided here

The obvious way to get a number from 1 to 49 is to take a random 32-bit value and use the remainder after dividing by 49. That is subtly wrong: 2³² is not a multiple of 49, so the first few numbers in the range come up very slightly more often. The bias is tiny but it is real and it is systematic. This generator rejects and redraws values that fall in the uneven tail, which removes the bias entirely at the cost of an occasional extra draw.

Unique draws versus independent draws

With duplicates allowed, each number is an independent event — rolling 6 twice in a row is unremarkable. With duplicates disabled you are dealing from a deck: each pick reduces the pool, and the draw becomes a combination rather than a sequence of independent events. Lottery draws, raffle winners and sampling without replacement all need the unique mode; dice and coin flips need the other.

Randomness does not look random

Genuine random draws produce clusters, runs and gaps that feel wrong. Six random numbers from 1 to 49 will contain two adjacent values about half the time. People asked to make up random numbers spread them out far too evenly, which is precisely how fabricated data gets detected.

Frequently Asked Questions