Fair Division Tool
Distribute items fairly between people.
Built and maintained by Paul Clark, Redmoon Software · Last checked · Changes & corrections
The Fair Division Tool shuffles your items and deals them out in turn, so nobody gets more than one item more than anyone else. Use it for sharing chores, splitting loot, dividing tasks, or handing out shifts so no one feels shortchanged. Unlike a raffle, everyone walks away with a fair share rather than one winner taking all.
Enter. List the items to share and the people who will receive them.
Click Divide to see a result.
Options
Recent results
- No history yet.
How it works
- Enter. List the items to share and the people who will receive them.
- Divide. Click Divide to spread the items as evenly as possible.
- Review. See who got what. When the items don't divide evenly, the extras go to people chosen at random, one each.
- Redo. Divide again for a fresh allocation if you want a different split.
About the Fair Division Tool
How the items are dealt
The item list is shuffled and then dealt round-robin across the people — one item each, wrapping around until the items run out. That guarantees the counts are as even as arithmetic allows: with eight items and three people, two people get three and one gets two.
Who gets the extra is random as well. Before every deal the order of the people is shuffled too, so when the division is uneven the surplus goes to people chosen at random, not to whoever is at the top of your list. The list order only decides how the shares are displayed.
Equal counts are not equal value
This divides by number of items, and nothing else. Three items each is only fair if the items are comparable — and for the cases people most often reach for a division tool, like splitting an estate or dividing chores, they are conspicuously not.
For genuinely unequal items, the classical answer is not a random deal at all: it is a procedure like "one person divides, the other chooses", or taking turns picking. Use this tool when the items really are interchangeable, and a turn-based method when they are not. There is more on the uneven case here.
What the output looks like
- Both lists are saved in this browser as you type, so a reload does not lose your setup.
- There is no seed, so a division cannot be reproduced — copy the result if it matters.
- The result is grouped per person, which is the form you want to paste into a message.
- Because shares are listed by name, each person must appear once — two entries called Alex (capitals ignored) are refused rather than merged. Add an initial to tell them apart.
- Only the last ten divisions are kept.
Frequently asked questions
What if the items don't divide evenly?
Counts differ by at most one, and who gets the extras is random too: before each deal the tool shuffles the order it deals to people in, so with 17 items and 4 people each person has the same 1-in-4 chance of getting the fifth. Your People list only sets the order the shares are shown in.
How is this different from a raffle?
A raffle gives everything to one winner. Fair division spreads the items so everyone gets at least one when there are enough to go around.
Is the random assignment unbiased?
Yes. Both shuffles, the items and the order people are dealt to, use your browser's Web Crypto API, so which items each person receives and who gets any extras are both unbiased.
Is my list of items and people private?
Yes. Everything is processed in your browser and works offline, so nothing you enter is sent to a server.
Related randomizers
Further reading
- Dividing things up when the numbers don't come out even
Splitting 17 items between 4 people means somebody gets a fifth one. Here is how a fair division tool deals items out at random, why it splits by count rather than value, and how it decides who ends up with the extras.
- Who gets which prize? Let chance decide — and show your work
Picking winners is the easy half of a prize draw. The fight starts when someone has to decide who gets the good prize and who gets the mug. A random prize assigner matches people to prizes in one shuffle nobody chose.
- A chore roster that actually feels fair (and stops the arguments)
The same person always ends up on bathroom duty. Here's how a random chore roster — freshly shuffled every day — spreads the jobs nobody wants without anyone keeping score, and how it handles having more chores than people.
- Matching two lists fairly: zip prizes, tasks, and partners at random
Assigning a list of people to a list of prizes, chores, or projects sounds trivial until you do it by hand and someone notices a pattern. Here's how a two-list matcher zips one list to a shuffled copy of another, what happens when the lists are different lengths, and when to reach for a different tool.