Random Picker
Draw names or numbers from a random source whose fairness can be argued, not merely asserted.
Runs in your browserYour list
Add a weight after an asterisk: Alice *3 makes Alice three times as likely. Weights are rounded to whole numbers, minimum 1.
Draw settings
of entries still in the pool, already drawn
Result
About this tool
When a draw decides something — who presents first, who gets the last slot — the interesting question is not whether the tool is random but whether it is fair, and fairness is something you can reason about. This page draws its random values from the browser's cryptographic generator, and it rejects the values that would tilt the distribution rather than taking a shortcut that quietly favours the low numbers. It also supports weights, so "twice as likely" is possible, and it turns those weights into whole numbers first, because that is what makes the result exactly proportional and open to checking. Lists, number ranges, draws without replacement and a shuffle are all here, and nothing leaves the page.
How to use it
- 1Paste your list, one entry per line. Add a weight after an asterisk if some entries should come up more often — Alice *3 makes Alice three times as likely as an unweighted entry.
- 2Choose how to draw: a single entry, several at once, or a full shuffle. With "do not draw the same entry twice" on, entries are removed from the pool as they come up, which is what you want for picking several people from one list.
- 3Press Draw. The result appears with the entries that are still in the pool noted underneath, and the session history keeps a record of what has already come up so you can prove the draw was run once and not repeated.
Frequently asked questions
Is this really random, and how would I know?
It uses the browser's cryptographic random generator rather than a seeded sequence, and then it discards part of what that generator produces. That second step is the one worth explaining: asking for a number and taking the remainder can bias the result whenever the generator's range is not a multiple of the number of options, so the honest approach is to throw away the values that would cause the bias and ask again. That is what happens here. Nothing about the draw is recorded or reproducible, so the same list does not produce the same result twice.
How do the weights work, and why whole numbers?
A weight of 3 means that entry is three times as likely as an entry with the default weight of 1. Weights are rounded to whole numbers, with a minimum of 1, and that is deliberate: with whole-number weights the draw becomes a single uniform pick from a total that is exactly the sum of the weights, and each entry's chance is its weight divided by that total — a proportion you can work out by hand and check. Decimal weights would produce results that are not exactly proportional, and there would be no way for you to tell from the outside.
Why were two identical names kept separate instead of merged?
Because two identical entries are far more often two people with the same name than one person listed twice, and merging them would silently give one person no chance of being picked at all — the exact opposite of what a draw is for. So entries are tracked by their position in the list, not by their text, and two names that read the same are two separate entries. The count of repeated entries is shown under the text area so you can see that it has happened and remove a genuine duplicate yourself.