Rank stability

Weighted prioritisation scores are guesses wearing a suit. This runs your scoring thousands of times with the weights and scores wobbled, and tells you which ranks survive — and which are noise you shouldn't argue about.

Try:
Scores 1–10. Weighted benefits ÷ effort (WSJF-style). Rename criteria and weights to match your scheme.

One row per initiative: name, then 4 numbers (Value, Time criticality, Risk reduction, Effort). Tabs, commas, or markdown-table pipes all work — paste straight from Excel, Sheets, or a doc. Replaces the current table.

Add initiatives and scores above. The result isn't a ranking — it's how much of the ranking you can trust.

What survives the wobble

median rank 90% rank range P(top-k) = share of simulations where the initiative makes the cut
Wobble: weights ±% · scores ± pts · 4,000 simulations · deterministic

Why wobble the weights?

Frameworks like RICE and WSJF produce a ranked list from scores nobody believes to one decimal place. The ranking inherits that false precision — and teams burn hours debating whether the weight on "time criticality" should be 2 or 3, when the honest question is whether the answer would change either way. This tool asks exactly that: it re-runs your scoring 4,000 times with every weight perturbed within ±50% and every score within ±1, then reports how often each initiative lands where. Ranks that hold across the wobble are real signal; ranks that reshuffle are ties — decide those on strategy, sequencing, or gut, and stop pretending the spreadsheet decided.

The idea is plain sensitivity analysis applied to decision frameworks — see Douglas Hubbard's How to Measure Anything on modelling what you're unsure about, and Itamar Gilad's critique of scoring prioritisation for why point estimates mislead.