Estimated scent match. We use the dedicated similarity percentage in a published Chaos clone guide for the exact fragrance edition. The newest matching Clones & Alternatives guide is the source of truth. An approved dupe-of comparison score is used only when no matching clone-guide entry exists, converted from 0–10 to a percentage. We do not generate percentages from guide rank, overall ratings, note overlap or buying recommendations.
Editorial evidence. Strong means an estimate with all three scent topics, or at least two topics plus linked coverage in both Clones & Alternatives and a dupe-of guide. Moderate means two topics, or an estimate with one topic or both linked guides. Limited means less coverage. Linked percentages are counted once. This describes our editorial support, not laboratory accuracy or independent consensus.
Guide order. This follows the Chaos guide covering the most available alternatives for this original, using its explicit ranks or otherwise its displayed order. A guide’s overall ranking can consider value and wearability; the highest-ranked bottle need not have the highest scent-match estimate. Guide position also breaks equal-estimate ties within that guide. Unscored bottles remain available to compare.
General consensus. Automated review checks extracted pages for the exact editions and quoted support. Firsthand comparisons carry more weight; duplicate discussions and authors count once, and each site’s influence is capped. Three distinct comparisons across two sites are required to adjust the score. Explicit resemblance percentages are used when quoted; otherwise very close, close, partial and weak scent comparisons map to 96, 87, 70 and 45 as internal weighting signals, not published reviewer ratings. External influence is at most 25%, reduced for conflicting evidence, with a five-point adjustment cap. Missing Chaos scores stay unscored. Excluded, changed or expired evidence is withdrawn until reviewed again. This adjustment applies only in Clone Lab.