Random Number Generator 1-1000

When a hat won’t do

Up to about fifty entries, you can draw a winner almost any way you like. Past a few hundred, physical methods fall apart. Writing out 800 slips of paper takes an afternoon, and nobody believes a hat with 800 slips was shuffled properly. That’s the problem this page solves: a big, fair draw where every entry has exactly a 0.1% chance.

The setting it’s built for is the online giveaway. A creator posts “comment to win,” gets a few hundred replies, and needs to pick one. Here’s the process we’d follow, and it’s mostly about keeping a paper trail.

A giveaway method you can defend

1. Close the entries at a stated time. Put the deadline in the original post, with a time zone. Late comments don’t count, and you shouldn’t have to argue about it.
2. Export or copy the entry list. For YouTube, our comment picker takes pasted comments and does the whole draw for you. For other platforms, copy entries into a spreadsheet so each one sits on a numbered row.
3. Remove ineligible entries before drawing. Duplicates (if your rules say one entry per person), your own account, anyone outside the eligible region. Do this first, then renumber. Removing people after you’ve seen the winner looks dodgy even when it isn’t.
4. Set the range to match the list. If you have exactly 1,000 eligible rows, this page is perfect. If you have 612, use the main generator with 1 to 612.
5. Record the draw. A screen recording of you clicking Generate, with the spreadsheet visible, is the gold standard. A screenshot of the result with the history panel showing is the minimum.
6. Draw backups. Generate three or four numbers with No repeats ticked. If the first winner doesn’t reply within your stated window, move to the next number on the list. You’ve already drawn them, so nobody can accuse you of picking a friend second time round.

A note on the rules

Giveaway law varies a lot between countries, and we’re not lawyers. One general point holds in many places, including most US states: a promotion where people must pay to enter, there’s a prize, and the winner is chosen by chance can count as a lottery, which is heavily regulated. That’s why so many sweepstakes say “no purchase necessary.” If money changes hands for entries, check your local rules before you draw.

What a thousand numbers look like on a wheel

The wheel below the generator still works, but with 1,000 slices each one is 0.36 degrees wide, thinner than the line between them on most screens. You’ll see a blur of colour and no labels; the winning number appears in the result card. It’s fun to watch on a stream. For anything you need to record, the plain generator is clearer.

The numbers at this scale

Probability facts for random numbers from 1 to 1,000

Fact	Value
Possible results	1,000 (1 to 1,000, inclusive)
Chance of any one number	1 in 1,000 = 0.1%
Even numbers / odd numbers	500 even (50%) · 500 odd (50%)
Prime numbers in range	168 (16.8% chance of a prime)
Same number twice in 2 picks (repeats on)	1 in 1,000 = 0.1%
A chosen number appears at least once in 10 picks	0.996%
Picks until a repeat is more likely than not	38 picks
Average (expected) result	500.5
Average picks to hit one specific number	1,000

Two rows stand out. The chance that a specific ticket wins in a single draw is 0.1%, so someone holding ten tickets has a 1% chance. And repeats show up much sooner than intuition says: with repeats switched on, you only need about 38 picks before a repeat is more likely than not. That’s the birthday problem again, and it’s one more reason to keep No repeats ticked for multi-winner draws.

Sampling a big list

The second big job for 1 to 1,000 is sampling. Auditors, quality checkers and researchers often can’t look at every record, so they check a random handful. If you have an invoice ledger with 1,000 lines, generating 30 numbers with No repeats and checking those 30 lines gives you a sample nobody chose by hand.

Sampling method	Who chooses the records?	Risk
First 30 rows	The order of the file	Early records are often unusual (the start of a period, test entries)
Records you “think look interesting”	You	Your hunch shapes the result
Every 33rd row	A fixed rule	Can line up with a pattern in the data, like weekly cycles
30 random row numbers	The generator	Nobody, which is the point

The RAND Corporation took this so seriously that in 1955 it published a whole book called A Million Random Digits with 100,000 Normal Deviates, so researchers had a trusted source of random numbers before computers were common. It’s still in print. This page is the faster version.

Test data and odd jobs

• Fake order numbers for a spreadsheet template or a demo.
• Random page numbers in a long book for a reading challenge. War and Peace has well over a thousand pages in most English editions, so you’ll need the main generator for that one.
• Seat numbers in a big auditorium for a surprise prize under the chair.
• Lucky draw at a conference. Badge numbers up to 1,000, drawn at the closing session.

If your list has names rather than numbers, paste it straight into the wheel spinner instead; it holds up to 1,000 entries and announces the winner by name.

How long does a big draw take?

About one second. That’s the real advantage at this size. A paper draw of 1,000 tickets takes a team of volunteers most of an evening to prepare. Here, the slowest part of the whole process is checking the winning number against your list, which is why we suggest putting entries in a numbered spreadsheet first. Sort by row number, scroll to the winner, done.

One last practical point: before a live draw, generate a couple of test numbers off-camera to check your range is set correctly. It’s an easy way to avoid the embarrassment of drawing number 1,000 from a list that ends at 640.