Elo Ranker
Rank items by simple pairwise comparisons (for example: career values, project ideas, or personal goals); the tool uses the Elo rating system to compute your final preferences.
The question used for each pairwise choice. Wording can bias results: positive framing highlights gains; negative framing highlights risks and losses.
How much a single choice updates the ratings. Higher values make ratings converge faster.
The number of pairwise decisions you will make. More rounds lead to more precise ranking.
Which do you value more?
Methodology
Elo rating system
Each item starts at a rating of 1000. When you choose item A over item B, the expected score of A is:
E_A = 1 / (1 + 10^((R_B - R_A) / 400))
Then ratings update:
R_A' = R_A + K(1 - E_A)
R_B' = R_B - K(1 - E_A)
K-factor used: . Higher K means faster convergence, suitable for deterministic preference judgements.
Pairing strategy
The first comparisons use a random complete covering (each item appears once) to seed initial ratings. Subsequent rounds use Swiss-tournament pairing: items sorted by rating, then adjacent (gap 1) and near-adjacent (gap 2-3) pairs are generated. This concentrates effort on decision boundaries.
Transitivity
Elo produces a strict total ordering. Swiss pairing refines adjacent boundaries, so the final ranking is transitively consistent. With comparisons for items (~x the item count), convergence is reliable.