WheelRoost Always free

The wheelroost notebook

Slice Weights, Computed: What Doubling One Slice Really Does

Every weighted-wheel tutorial says some version of "give your favorite slice a bigger share." Nobody prints the share. So we took the same weighted-sampling engine that runs every spin on this site, put one slice of an 8-slice wheel on weights from ×2 up to the engine's ×50 cap, and tabulated the probability per spin, the odds of hitting it within three spins, and how many spins you need for a coin-flip or a near-certainty.

By Nan Liang
October 28, 2026 · 6 min read

Every number below comes out of the same two functions that power a real spin here — weightedPick does the draw and mulberry32 is the seeded random source — plus exact binomial math for the multi-spin columns. The last column is the engine itself: 200,000 seeded draws per row, no shortcuts.

The table: one weighted slice on an 8-slice wheel

Slice weightOdds per spinHits ≥1 time in 3 spinsSpins to 50%Spins to 90%Engine check (200k draws)
×1 (baseline)12.5%33.0%61812.6%
×222.2%52.9%31022.3%
×330.0%65.7%2730.0%
×541.7%80.2%2541.8%
×1058.8%93.0%1359.0%
×50 (engine cap)87.7%99.8%1287.7%

Baseline is 1/8 = 12.5% per spin. The ×2 row is the one people actually use: 22.2% per spin, a coin-flip in 3 spins, a near-lock in 10. The "spins to" columns are exact: n = ⌈ln(1−target) ÷ ln(1−p)⌉. The engine check column is a Monte Carlo run of the site's own sampler — every row lands within ±0.4 points of theory, which is what 200,000 draws should do.

Three readings of the table

Doubling is the sweet spot, and it is bigger than it feels. Going ×2 lifts one slice from 12.5% to 22.2% — nearly double the share, because your slice now carries 2 of the 9 total shares, not 2 of 16. In a classroom of "who presents first," that is the difference between a gentle nudge and a loaded die.

The curve is front-loaded, then it flattens. ×5 already gives you an 80% chance of at least one hit in just three spins; ×10 buys 93%. Past that, extra weight mostly squeezes the tail: ×50 versus ×10 changes three-spin odds by only ~7 points. If a wheel ever feels "broken," check for a stray ×10 before blaming the spinner.

Weight is relative, always. The denominator is the sum of all slice weights, so adding a second heavy slice quietly dilutes the first. Two slices at ×3 on an 8-slice wheel each hold 3/13 ≈ 23% — barely above a solo ×2. The name picker shows the ×N badge next to every weighted slice precisely so this arithmetic is never hidden.

Try it on your own wheel

  1. Open any wheel — the NBA team wheel is a good lab — and expand Edit slices.
  2. Add a comma and a number after one line, e.g. "Boston Celtics, 2", and apply.
  3. Spin a few times, then check the slice list: the badge shows the exact weight the engine is using.
  4. For a fair draw, delete the weight. For a staged moment, that is what presenter mode is for — labeled, visible, and separate from the odds.
How do I set a slice weight on my wheel?
Open any wheel, expand Edit slices, and put the weight after a comma on that line — "Pizza, 3" gives Pizza three shares. Weights above 50 are clamped to 50, and anything that is not a positive number falls back to 1, same rules the sampling engine enforces.
Is weighting a slice the same as presenter mode?
No — they solve opposite problems. Weights are an honest, visible skew: the slice list shows a ×N badge, and every spin is still a genuine weighted draw. Presenter mode stages a rehearsed moment and is labeled as such in the editor. If the group expects chance, use weights or nothing; if the group expects a show, that is what presenter mode is for.
Why is the ×50 row not 50 times the base odds?
The ×50 cap exists so one slice can dominate without mathematically erasing the rest. On an 8-slice wheel the weighted slice holds 50 of 57 total shares — 87.7% — because the other seven slices still carry 1 share each. Weight is relative to the sum of all slices, not a multiplier on probability.
Do weights change after elimination removes a slice?
Yes, and correctly so. Elimination removes the picked slice and its shares from the pool, so the remaining weights renormalize automatically. A ×3 slice among six remaining slices holds 3 of the remaining total, not 3 of the original eight.

← All posts