Baby Boom data and methodology

Where Baby Boom’s names come from: Social Security card application data since 1880, and how each name’s peak year is measured.

Where the names come from

Every name in Baby Boom comes from data the U.S. Social Security Administration publishes itself: a full national count of babies given each name, by sex and by year, drawn from Social Security card applications going back to 1880. The Administration only publishes a name for a given sex and year once at least 5 babies received it — a privacy floor built into the source data itself, not something Baby Boom adds on top. Very rare or one-off names are filtered out before Baby Boom ever sees them.

Share, not the number of babies

A name’s peak year is the year its share of that sex’s births was highest — its count divided by the total recorded for that sex that year, meaning every name published for it, not literally every baby of that sex born — not the year it collected the most babies in absolute terms. The distinction matters because U.S. births themselves rose sharply from 1946 to 1964, the actual baby boom. Ranking by raw count would drag almost every name’s “peak” toward those years, simply because the population of babies being named was so much larger then. Measuring share instead compares each year on its own terms, so a name’s peak reflects how fashionable it actually was among the babies born that year, not how many babies happened to be born at all.

How the pool is chosen

Not every name the Social Security Administration has ever published belongs in the game. Baby Boom’s pool is limited to names that were, in their peak year, among the 350 most common for their sex, and that show a clear rise and fall: at its peak, a name’s share was at least four times what it had been thirty years earlier, and at least four times what it was thirty years later. A peak less than thirty years from the end of the data — after 1995 — has no year that far ahead to check, so for those names the rule instead requires 2025’s share to be at most two-thirds of the peak share. Names still climbing in the last few years of data are left out entirely, since their real peak has not happened yet. Applying those rules to the full history leaves 1,185 names: 775 girls’ names and 410 boys’ names.

Records begin in 1880

The Social Security Administration’s national data starts in 1880, so a name that was already declining before recordkeeping began has no earlier year to reveal an even higher share. Its peak is recorded as 1880 itself, the first year available, rather than the true year — some point before 1880 — when the name was actually most popular. Cora is one such name: her recorded peak is 1880, but the share was almost certainly higher at some earlier point the data cannot see.

Girls’ names and boys’ names

Baby Boom sorts every name into “girls’ name” or “boys’ name”, following the Social Security Administration’s own two published categories. A name given to both sexes, such as Peyton, is tracked as two separate entries with two separate peak years — Peyton peaked as a girls’ name in 2009 and as a boys’ name in 2007 — so a single spelling can appear in the pool twice, once for each sex, each with its own history.

Updates and fairness

The five names chosen for a published puzzle never change once that puzzle is live. The Social Security Administration publishes a new year of national data once annually, each May, and Baby Boom’s dataset is refreshed on that same yearly cycle; any correction needed for a name that appears in a future puzzle is made before that day. To report a data problem, email hello@guessday.com with the puzzle number.

Attribution and independence

Baby name data is published by the U.S. Social Security Administration, a United States government agency, and is in the public domain. Guessday is not affiliated with or endorsed by the Social Security Administration. The data describes naming patterns across the whole country, not any individual person, and nothing on this site is legal, medical or government guidance.