# U.S. baby names: the shrinking share of the top 10

Prepared from the reviewed SSA snapshot retrieved 2026-10-14T11:14:00Z.

## Result
In SSA's national records for birth year 1950, the year's ten most popular boys' names covered 605,169 of 1,818,762 boys (33.27367737%). For 2025 this was 136,355 of 1,835,823 (7.42745897%).
For girls the corresponding values were 412,876 / 1,758,531 (23.47846015%) and 112,268 / 1,757,924 (6.38639668%).
The chart uses every annual observation from 1950 through 2025. Its percentage axis starts at zero and runs to 40%; the top 10 is selected independently for each sex and year.

## Files
- baby_names_top_10_share_1950_2025.png: 1800 × 1500 pixel post image.
- baby_names_top_10_share_1950_2025.csv: all 152 plotted observations, with exact numerators, denominators, and ten names.
- ssa_baby_names_1880_2025.csv: 2,181,032 published name/sex/year records, standardized into a single long-form CSV.
- ssa_annual_applicant_totals_1880_2025.csv: total applicants, published counts, unpublished remainder, published distinct names, and top-1000 counts/shares.
- build_baby_names.py: cleaning, validation, chart renderer, and bundle creation.
- manifest.json: retrieval time, source URLs, checksums, endpoint calculations, and checks.
- raw/: original national ZIP and applicant-totals page used for the build.

## Source and provenance
Social Security Administration, Office of the Chief Actuary, Popular Baby Names.
National ZIP: https://www.ssa.gov/oact/babynames/names.zip
Download landing page: https://www.ssa.gov/oact/babynames/limits.html
Complete annual applicant denominators: https://www.ssa.gov/oact/babynames/numberUSbirths.html
Source qualifications: https://www.ssa.gov/oact/babynames/background.html
The published source describes records as of March 2026. Retrieved on 2026-10-14T11:14:00Z.
The ZIP includes SSA's original NationalReadMe.pdf. Earlier-year counts can be revised in later SSA snapshots.

## Public-domain status
SSA states that statistics and documents prepared by its staff in their official federal work are in the public domain: https://www.ssa.gov/policy/accessibility.html
SSA's official Data.gov record for this exact national ZIP explicitly lists CC0 1.0: https://catalog.data.gov/dataset/baby-names-from-social-security-card-applications-national-data
License: https://creativecommons.org/publicdomain/zero/1.0/
The original government data are also public domain in the United States.
The cleaned tables retain those factual records. No additional restrictions are asserted over the cleaned factual data.
No personal identifiers, row-level birth records, or private data are included; counts are public aggregates.
No endorsement by SSA is implied.

## Cleaning and calculation
Input national files are named yobYYYY.txt and contain name, sex, count. All published national rows from 1880 to 2025 are retained. No spelling variants are merged and no gender inference is made. The source's M/F sex categories are retained.
Within each year/sex, names are sorted by descending count, then alphabetically to break ties, matching SSA's rank convention.
Top-10 share = 100 × sum of counts of the ten highest-ranked names / SSA's complete applicant total for the same sex and birth year.
The denominator is NOT the sum of the downloadable names. SSA suppresses names with fewer than five occurrences, so summing the public file would overstate top-name shares.
Published-name counts and the unpublished remainder are stored as validation fields. The latter is an aggregate difference, not reconstructed names or counts for particular suppressed names.

## Coverage and limitations
- These are U.S.-born Social Security card applicants, not a census or CDC estimate of all live births. The title's baby-name shorthand is qualified on the chart and in the source comment.
- National data cover the 50 states and Washington, D.C.; territories are excluded.
- SSA includes records with known birth year, sex and state of birth and a first name of at least two characters.
- Many people born before 1937 never applied for Social Security cards. The full dataset preserves early years, but the chart begins in 1950 to avoid the most severe early coverage limitations. This does not make any year a literal census of births.
- Names with fewer than five occurrences are omitted from public name files. All applicant totals remain the denominator.
- Spelling variants are distinct. SSA removes spaces and hyphens, so this measures concentration in recorded spellings rather than a count of underlying name meanings or pronunciations.
- Source entries can include recording errors or placeholder names; none are silently removed.
- The chart shows concentration among the leading ten names. It does not measure all aspects of name diversity, demonstrate causation, or describe living adults' current names.
- Rank and percentage values are computed fields; count and denominator values remain integer source values. CSV percentages are rounded to eight decimal places; the plot uses full precision.

## Reproduce
Python 3.10+ and Pillow. Install Pillow from PyPI in your own environment.
Run: python build_baby_names.py --output output
For an exact repeat from the bundled source snapshot:
python build_baby_names.py --raw-dir raw --output output
The script uses DejaVu Sans and DejaVu Sans Bold. Install those fonts or place the .ttf files in a fonts/ folder beside the command's working directory.
Checks assert source-row retention, key uniqueness, minimum counts, valid coverage, denominator consistency, endpoint totals, and 2025 top-1000 percentages against SSA's published coverage table.

## Download formats
The chart CSV covers 1950–2025. The full name CSV and applicant-total CSV cover
1880–2025. The full name CSV exceeds the hosting limit when uncompressed, so
its separate download uses lossless gzip compression (.csv.gz). Decompress it
with a gzip-capable archive utility, or read it with a data tool that supports
gzip. The complete ZIP includes the original uncompressed CSV, all source
snapshots, the approved chart and reproduction code. No rows or precision were
removed to fit the hosting limit.

## Field reference
- year / sex / name / count: original SSA birth-year, M/F category, spelling and
  integer count. rank_within_sex is computed; share_of_sex_percent uses the
  complete same-year, same-sex applicant denominator.
- top_10_count / ssa_applicant_total / top_10_share_percent / top_10_names:
  exact top-ten numerator, complete denominator, percentage rounded to eight
  decimals and semicolon-separated names for each chart observation.
- published_name_count / unpublished_name_count / published_distinct_names:
  total published count, aggregate denominator remainder and published spellings.
- top_1000_count / top_1000_share_percent: the leading thousand's count and share.

## Publisher verification
All 2,181,032 name records were checked against the bundled national raw ZIP.
All 292 year/sex applicant totals match the raw denominator page; all 152 chart
observations, ranks and derived shares reconcile. No source errors were corrected,
names merged, rows deduplicated or missing/suppressed names imputed for this release.
The supplied CSVs and chart are unchanged. manifest.json retains the source
pipeline's checks; RELEASE-MANIFEST.json records the actual release package.

Dataset page: https://meplar.com/datasets/us-baby-names/