What is the most random number?
Mathematically, no number is more random than another: in a fair draw every number has exactly the same chance. But when people are asked to pick a "random" number, they choose some far more often than others. Between 1 and 10 the favourite is 7; between 1 and 100, numbers like 37 and 73 stand out. For a truly fair pick, use the random number generator.
Randomness is a property of the process
"Random" describes how a number was chosen, not the number itself. A fair die is random because each face has a 1 in 6 chance, even if it lands on 6 three times in a row. The number 7 is not random or non-random on its own; it is random if it came out of a fair process.
That is why the question "what is the most random number" has no mathematical answer. It does have a psychological one.
Why people pick 7
Ask a group of people to pick a number from 1 to 10 and 7 wins by a wide margin in many informal surveys. A few reasons are often suggested:
- Extremes feel planned. 1 and 10 feel too obvious, so people avoid the edges.
- Even numbers feel tidy. 2, 4, 6, 8 and 10 seem less "random" than odd numbers.
- 5 is the middle. It feels too neat as well.
- 7 is a prime and culturally loaded, from seven days of the week to "lucky 7".
What remains is 3 and 7, and 7 tends to win.
Between 1 and 100: 37 and 73
The same instincts work on a larger scale. People avoid multiples of 10, numbers with repeated digits (like 44 or 77, although 77 is popular in some surveys) and round-feeling numbers like 25 or 50. Numbers ending in 7, and primes such as 37 and 73, are picked far more often than chance would predict.
People are bad random number generators
Humans are also poor at producing random sequences. Asked to write down 20 coin flips, most people switch between heads and tails too often and avoid streaks. In a real sequence of 20 fair coin flips, a run of four or more of the same side is more likely than not. Our sense of "random" is closer to "evenly spread", which is something quite different.
Real randomness
Computers usually generate pseudo-random numbers: a formula produces a sequence that looks random but is fully determined by its starting value. For games and simulations, that is fine. For passwords, lotteries and security, you need a cryptographically secure generator, which mixes in unpredictable data from the operating system.
The random number generator on this site uses the browser's crypto.getRandomValues(), a cryptographically secure source, and avoids "modulo bias" so every number in the range is exactly equally likely. Nothing is stored or sent.
Fair draws in practice
- Raffles and giveaways: number the entries and draw one number between 1 and the number of entries, ideally with the draw visible to everyone.
- Picking teams or turns: generate several numbers without duplicates.
- Board games: a virtual die is simply a random integer from 1 to 6.
Frequently asked questions
What is the most common random number between 1 and 10?
When people pick, 7 is chosen most often. For a fair generator, every number from 1 to 10 has exactly a 10% chance.
Is 17 the most random number?
17 is often named as a "random-feeling" number between 1 and 20, for the same reasons as 7: it is odd, prime and not near the edges. Mathematically it is no more random than any other number.
Can a computer generate truly random numbers?
Ordinary functions produce pseudo-random numbers. Cryptographic generators combine these with unpredictable input from hardware and the operating system, which is random enough for security purposes.
Is a streak a sign that a generator is broken?
No. Streaks are normal in random data. A generator that never produced streaks would actually be suspicious.