Using a Wheel as a Decision Maker

Handing a decision to a wheel sounds like giving up. Sometimes it is exactly the right move, and there is a clean test for which situation you are in: compare the cost of deciding with the difference between the options.

Six people spending twenty minutes choosing a restaurant have spent two hours of collective attention on a difference of maybe fifteen minutes of enjoyment. The deliberation costs more than the decision is worth. That is not a case for deciding badly. It is a case for deciding quickly, and a wheel is the fastest way to turn a shortlist into an answer without anybody having to own it.

The corollary matters just as much. When the options genuinely differ, randomising is not neutral. It is choosing to ignore information you have. Knowing which kind of decision is in front of you is the whole skill.

The three conditions

A decision suits a wheel when all three of these hold. If any one fails, work the problem instead.

Every option is acceptable. Not equally good, just acceptable. You would shrug at any outcome. If one result would genuinely annoy you, it does not belong on the wheel.
Deliberating further is not producing new information. Going round the same three points for the fourth time is not analysis, it is stalling. If a five-minute check would actually settle it, do the check.
The decision is reversible, or small. Dinner is reversible. A job offer is not.

Weighting: you do not have to pretend to be indifferent

This is the part people miss. A coin flip forces a 50/50 you may not feel. A wheel lets you say “I lean towards Thai, but I am open to the others” and turn that into actual odds.

Give Thai a weight of 3 against two other options at 1 each and it wins 60% of the time while the alternatives stay live. That is a genuinely different statement from either picking Thai outright or flipping a fair coin, and it is often the honest description of how you feel. The worked arithmetic is here. On the free wheel you get the same effect by listing an option more than once.

There is a useful side effect. Deciding what weight to give each option forces you to state your preferences numerically, and that alone frequently resolves the decision before you spin. If you find yourself giving one option a weight of 20 against everything else at 1, you have not built a wheel. You have discovered that you already decided.

The wheel as a preference detector

The most useful thing about spinning is what happens in the second after it lands. If the result produces relief, the decision was genuinely close and you are done. If it produces a flinch, a small internal “no, not that one”, you have just learned that you had a preference you were not admitting to.

Act on the flinch. Nothing obliges you to follow a wheel you asked for advice from, and a spin that surfaces a hidden preference has done more for you than one you obeyed. This is the main reason to use a wheel on a genuinely difficult personal decision: not to outsource the answer, but to provoke a reaction you can read.

It follows that you should spin privately for anything you might want to overrule. A public spin creates an expectation that the result stands, which is the right property for a raffle and the wrong one for working out how you feel.

Group decisions and blame

In a group, the wheel does something a decision cannot: it removes the decider. Nobody chose the restaurant, so nobody is responsible for it being mediocre, and the person who would otherwise have made the call does not spend the evening watching for signs of disappointment.

For this to work, agree two things before spinning:

Everyone confirms every option is acceptable. One veto removes an option, no explanation required. This is the step that makes the rest work, because a wheel containing an option somebody hated is a fight waiting to happen.
The result stands. Say it out loud. A group that treats the outcome as a suggestion has spent the deliberation time and got nothing for it.

Where a group is deciding something with any stakes, spin somewhere everyone can see it. The free wheel on a shared screen is fine. A shared link is better when people are remote, because everyone watches the same spin land rather than being told what it landed on.

Turn order and who goes first

Board games, presentation slots, who picks the film this week, who gets the good chair. Order confers a real advantage in most games and a real cost in most meetings, so “youngest first” or “whoever set up the board” is a rule that quietly favours the same person forever.

Draw the whole order in one pass with elimination on. It takes fifteen seconds, it settles the argument, and for anything recurring the draw history tells you who went first last time, which is usually the actual point of contention.

Chores and rotations

Elimination mode is the important setting for anything that should cycle. Draw one name per task until the wheel is empty and everybody has exactly one job, then reset and draw again next week.

Two refinements worth knowing. Draw the worst job first, because that is the one people care about, and drawing it while every name is still in the wheel is visibly fairer than leaving it to whoever is left at the end. And if one task is genuinely more unpleasant than the others, keep a record across weeks so nobody draws it twice in a row. A stored history does this without anybody having to remember.

What not to randomise

The line is not about how big the decision is. It is about whether the options differ in ways you can reason about.

Anything irreversible. Accepting an offer, ending something, spending money you cannot get back.
Anything affecting somebody who is not in the room. They did not agree that the options were all acceptable.
Medical, legal or financial choices. These have correct answers, or at least better and worse ones, and they need advice rather than chance.
Decisions you are avoiding because you know the answer. A wheel here is a way to make a choice you already made feel like it was not your fault. The flinch test catches this, because you will feel it land wrong.

A short procedure

Write out every option you would genuinely accept
Let anyone involved veto an option, no reason needed
If two options remain and one has any real edge, take it, because you have decided
Weight the rest to match how you actually feel
Agree out loud that the result stands, then spin
Notice your reaction. Relief means done; a flinch means you knew all along

Getting started

The wheel on the home page needs no account. Type the options in, list a favourite twice to weight it, and spin. For a shared link a group can watch together, or a saved wheel for a recurring rotation, see the how-to guide or create a free SpinForFree account.