Weights define proportions
Choices with weights 1, 2, and 3 create a total weight of 6. Their probabilities are 1/6, 2/6, and 3/6 because the wheel selects a random point across cumulative intervals. The numbers are ratios, so 10, 20, 30 produce the same chances as 1, 2, 3.
Tea, Coffee, and Water
For Tea = 1, Coffee = 2, and Water = 3, Water covers three of six weight units and has probability 50%; Coffee has 33.33%; Tea has 16.67%. In 60 independent spins the expected counts are 30, 20, and 10. The result of any one spin is not a quota.
Validate every weight
A zero weight creates no selectable interval; a negative value has no sensible probability here. The frequent error is calling a higher weight a score without explaining why it deserves greater chance. Record whether weights represent ticket counts, stock, or an agreed policy.
Weighted selection has a narrow job
This wheel does not produce a weighted ranking without repeats, balance outcomes over time, or make unequal treatment fair by itself. It is appropriate only when the unequal ratio is intentional and explainable.
Convert a policy to a table
Before spinning, write the labels, weights, total, and resulting percentages in a small table. For 1, 2, and 3 the total is 6, so the percentages are about 16.67, 33.33, and 50. This exposes a mistaken total immediately. It also distinguishes an intentional 3-to-1 priority from an arbitrary-looking number entered after somebody asks why one option is favoured.
Scaling does not change a ratio
A charity ticket draw with weights 4 and 12 gives the second label three times the first label's probability: 12 divided by 16 is 75%, while 4 divided by 16 is 25%. Rewriting those values as 1 and 3 changes nothing. This is useful for checking an entered rule: only relative sizes matter, so a claimed priority should survive division by the same positive factor.
Before applying unequal odds
Show the percentages implied by the weights and invite a check of the total. People can then agree with the probability policy itself rather than arguing only about whichever label appears.
Result record
Preserve the entered ratios with the outcome, because a later reviewer needs to see the chance rule as well as the selected label.