Weighted Random Picker
Pick items with custom probabilities.
Built and maintained by Paul Clark, Redmoon Software · Last checked
This picker lets you give each option its own weight so some entries come up more often than others. It's built for weighted raffles, loot tables, or any decision where the choices shouldn't be equally likely. Add a number after each item and the tool turns those weights into fair, proportional odds.
List. Enter each option on its own line, adding a weight after it where you want different odds.
0 items
Click Pick (weighted) to see a result.
Options
Recent results
- No history yet.
How it works
- List. Enter each option on its own line, adding a weight after it where you want different odds.
- Pick. Click "Pick (weighted)" to draw one option using those probabilities.
- Re-roll. Click again for another independent pick whenever you like.
About the Weighted Random Picker
Stating the odds instead of inheriting them
Each line can carry a weight, and an entry with weight five is five times as likely as one with weight one. A line with no weight is treated as weight one, so a plain list behaves exactly like an even draw.
This is not an unfair draw — it is a draw where the odds are written down. The alternative people usually reach for is pasting a name several times, which produces the same distribution but hides the odds from everyone reading the list. Explicit weights are auditable; repeated lines are not.
Drawing more than one, and what "unique" does
Set the count above one and, by default, each pick is an independent weighted draw — so a heavy entry can come up repeatedly, which is correct for sampling with replacement.
Tick Unique and the picks are distinct instead. The important detail is what happens to the odds: after an entry is drawn it is removed, and the remaining weights are renormalised over what is left. So the first pick follows your stated weights exactly, and later picks follow the weights of whatever survives. That is the standard behaviour for weighted sampling without replacement, and it means the second pick is not distributed the same way the first was.
Getting the weights right
- Weights are relative, not percentages. 3 and 1 is the same distribution as 30 and 10.
- A weight of zero excludes an entry entirely, which is a tidier way to suspend someone than deleting the line.
- Set a Seed for a draw you need to reproduce or prove.
- There is a longer piece on weighted draws, including when a weighted draw is the wrong model.
Frequently asked questions
How do I set a weight?
Add it after the item name with an x, comma, or colon — for example "Grand prize x1", "Carla, 5", or "Bob: 0.5". An item with no weight is treated as 1.
Does a higher weight mean more likely?
Yes. An item with weight 3 is three times as likely to be picked as one with weight 1.
Are the odds actually fair?
Yes. The pick is driven by your browser's Web Crypto randomness, so the long-run frequencies match the weights you set with no hidden bias.
Is my list of entries private?
Completely. Everything runs in your browser, and your items and weights are never uploaded anywhere.
Related randomizers
Further reading
- A snake draft evens the picks, not the totals
The Weighted Team Generator drafts serpentine — best player to team one, next to team two, then back again. It is a good heuristic and it is not an optimiser. Here is when it lands short, why re-rolling will not save you, and what to do instead.
- One tagged list, many decisions: designing a master list for filtered random picks
Most decision lists die because you end up with six overlapping copies — weeknight meals, weekend meals, takeout-only. Here's how to build one tagged master list instead, design a tag vocabulary that stays usable, and let a conditional picker slice it per occasion.
- Pick several names at once, not one slow draw at a time
When you need three winners or five volunteers, drawing them one by one invites doubt and double-takes. Here's how a multiple random picker deals a whole batch of unique names in a single click, and when to allow repeats.
- Weight a random draw so some options come up more often
Sometimes a fair coin flip is exactly the wrong tool — you want the grand prize to be rare and the consolation common. Here's how a weighted random picker lets you hand each option its own odds, with or without replacement.
- How to run a raffle or giveaway draw that everyone trusts
Pulling tickets from a hat looks fair but rarely is, and a public giveaway has to survive people who'll cry foul. Here's how to run a draw with weighted tickets, multiple prizes, and a seed anyone can re-check.
- Filter first, then pick: narrowing a mixed list before a random draw
One big list of options is convenient until the moment you only want a certain kind — a quick meal, an indoor activity, anything but spicy. Here's how tagging your items and filtering before the draw lets one list serve every mood, and the include/exclude rule that catches people out.
- Let a coin flip decide — but make it a smarter coin
Flipping a coin to make a decision isn't lazy — it's a useful tool for breaking analysis paralysis and surfacing what you actually want. Here's how weighted randomness helps you decide better, not just faster.