About

List Randomizer puts a list in an order nobody chose. Paste names, raffle entries, tasks or numbers, and it gives them back shuffled, draws winners with alternates, splits people into even groups or matches one list against another. Most people arrive with something to do next: call students up, announce a giveaway, form teams. The site is built so that the result is fair, believable and easy to take with you.

What it does

• Takes a pasted list, or opens one from an Excel (.xlsx), CSV or text file in the browser, letting you choose the columns that hold the names.
• Shuffles a list of up to 100,000 items, showing each item’s new place and where it came from.
• Draws any number of winners and, in the same draw, alternates in a fixed order.
• Splits a list into a chosen number of groups, or into groups of a chosen size, with a choice for the leftovers.
• Matches a list against a shuffled second list, row by row, and lists anything left over.
• Repeats any draw exactly from a seed, so others can check it.
• Copies, downloads CSV or text, makes share links that reopen the exact result, and presents one item at a time on a big screen.

Why I built it

Before writing any code, I put the same test lists through the ten list randomizers Google showed first for “list randomizer”. The eight I could re-run 300 times all shuffled fairly enough, but most of the ten damaged the list on the way. Four broke multi-word items: one turned every space into a comma, one split “Smith, John” in two, one split on every space so 1,000 lines came back as 2,000 words. One returned only 7 of 10 names because a hidden quantity box defaulted to 7. Eight said nothing when the list was empty or had one item, three rendered typed HTML as real markup, and only two kept the box and the button on a phone’s first screen. None could split a list into groups and draw winners in the same place, and none offered a share link that reopens the same order without signing up for an account.

How the shuffle works and how it was checked

The shuffle is Fisher–Yates (Durstenfeld’s version), fed by the browser’s cryptographic random generator, with rejection sampling so that no position is favoured. A seeded draw uses the sfc32 generator, started from a hash of the seed text. Automated tests run before every build: every item survives a 1,000-item shuffle exactly once, 60,000 shuffles of three items land within 5% of an even split across all six orders, the same seed always gives the same order, and the group sizes match hand-worked examples (25 people in 4 groups gives 7, 6, 6, 6). A separate run of 600,000 shuffles, quoted on the home page, gave each order of A, B, C between 99,844 and 100,204 times.

What it can’t do

It doesn’t read giveaway comments from social networks; you paste the entries. It can’t keep certain people apart or balance teams by skill, and it doesn’t handle Secret Santa, where nobody may draw themselves. It keeps no history between visits. For regulated lotteries, check whether your local rules require certified equipment or a witness.

Who runs it

The site is run by an independent developer, not a company. It has no advertisers or sponsors and no connection to any of the sites or apps mentioned on it. Lists never reach me: all of the work happens in your browser.

Statistics

The site counts page views and which buttons people use with its own analytics, never the items on your list. The privacy policy lists exactly what is recorded and has a switch to turn it off.

Corrections

If a result looks wrong, use the contact page and describe the mode, the settings and what you expected. A confirmed problem gets a permanent test so it can’t come back.