U.S. national SSA records, 1880–2025. Each spelling and SSA sex category is a separate record. We compare the same calendar years, so a 1950s peak does not become a match for a 1990s peak simply because both curves look alike.

Usage is measured per million published name records in the same category and year, not per million total births. This adjusts for changing dataset size; it is an approximation because unpublished names are excluded. Early records, particularly before 1937, are incomplete.

Our custom distance combines curve shape (45%), peak era (15%), recent momentum (15%), peak usage (15%) and current usage (10%). Turning off popularity level removes the last two factors and rescales the others. Curves use a centered five-year average (shorter at the ends); peak era is the highest smoothed usage year, with the latest year used for ties. Momentum compares the latest three-year average with the three-year average five years earlier, with a small stabilizer for rare names.

Similarity is 100 × (1 − weighted distance), rounded to a whole number. It is an exploratory score, not a probability, statistical significance test, or forecast. The closest available matches may still have meaningful differences.

To limit noise, names need at least 20 published years and a highest annual count of at least 50. Missing entries do not prove zero births: SSA withholds counts below five. For matching only, missing entries contribute zero as a lower-bound approximation. Charts show gaps for unpublished years. Rare-name matches are less reliable.