Weighted decision matrix
Compare your own 0–100 scores using normalized positive weights.
- Runs in your browser
- No upload
- Check before download
Names, notes and plans stay in this browser. No account, calendar connection or sending. Save only when you explicitly choose it.
Compare your own 0–100 scores using normalized positive weights. Sensitivity changes each weight ± the chosen percentage independently; ties remain visible, no recommendation. Criterion≤20 · Option2–30 · Score0–100
Questions
Limits:100rows (rosters64/50 as shown),256KiB backup,30,000text units,100,000search visits,20second worker deadline,40PDF pages/8MiB. PDF uses local Liberation Sans with actual glyph checks; no RTL/complex shaping. Unsupported print text remains in the backup. Removing a row removes its linked assignments; editing/reordering preserves positional identities. Roster/draw changes clear tournament results. Roster/reference names must be unique for generation; incomplete drafts may repeat them.
Device storage is unencrypted and may be cleared or seen by other users of this browser. Keep a portable backup. Load/Clear never contacts a service.
Liberation Sans · SIL OFL1.1@libpdf/core · MIT| Option | Synthetic sample1 | Synthetic sample2 |
|---|---|---|
| Synthetic sample A | ||
| Synthetic sample B |
How to use
- Add
Add criteria with positive weights, the options, and scores from 0–100, and select the sensitivity percentage.
- Create plan
Try this concrete example: weights[2,1],A[90,20],B[50,100] → A=B=200/3≈66.6667; weight0−25% → A62,B70
- Export backup
Look at ties and sensitivity, not just the highest total. Export the scores and keep the editable input backup.
Questions
Are weights normalized automatically?
Yes. Weighted scores are divided by the total weight. Sensitivity changes one criterion at a time, not every possible scenario.
What should I correct before generating?
Every option needs one score per criterion. Zero weights and scores outside 0–100 are rejected.
What does this output not establish?
A help for subjective scoring. It does not find an objectively best decision, model uncertainty or give investment advice.
Local PDF sheets use @libpdf/core0.5.1 (MIT) and Liberation Sans. Copyright notice and original license · Original Liberation font license (SIL OFL1.1)