Home › How the data is counted

How the data is counted

Every figure on this site comes from one dataset: the U.S. Social Security Administration's baby names data, as published through the Hugging Face dataset nkrishnaswami/us-ssa-baby-names-states (CC0, public-domain US federal data). The SSA's own download was not reachable from the build server, so the mirror is used and every published page is spot-checked against SSA's own top-5 table.

What the data covers

Births in the United States, 1910 to 2025, by state, sex and year. Each row is one name in one state in one year, with the number of babies given that name. The 2025 data was released by SSA in May 2026.

What counts, and what does not

SSA publishes only names given to at least 5 babies of one sex in one state in one year. Names given to fewer than 5 babies are not published and do not appear on this site. Exact spellings are counted separately: Mateo and Matteo are different names. Sex is as recorded on the Social Security card application.

How ranks are worked out

A rank is the name's position among published names for that state, sex and year, ordered by count. Names with the same count share a rank, and the next rank is skipped, which is why ranks are not always sequential.

How shares are worked out

When a page says "1 in every 214 girls", the denominator is the sum of the SSA-published counts for that state, sex and year. Because names given to fewer than 5 babies are not published, that sum is slightly smaller than the true number of births, so the share is a close estimate from published data, not an exact census figure.

What "only in Texas" means

A name counts as only popular in a state if it was given at least 5 times in that state in 2025 and fewer than 5 times in any other state. "Other states" means the 50 states and DC; Puerto Rico and the other territories are not counted as states for this. This definition follows the one published by nancy.cc, the only other site publishing these counts.

How rank changes are worked out

Rank change is the 2024 rank minus the 2025 rank, recomputed from the counts in the file. A name that was not published in 2024 has no rank change and is listed separately as new. On the national rising and falling pages, only names given to at least 100 babies in 2025 and at least 20 in 2024 qualify: without that rule the lists are topped by names jumping from 5 babies to 70, which is arithmetic noise, not a real trend.

Updates

The data is refreshed once a year, in May, when SSA releases the new year. Old-year pages are kept as dated archives and are never silently rewritten to a new year.

What the site does not do

No name meanings or origins (that data does not come from SSA). No middle-name or sibling-name suggestions (SSA does not publish middle names, and no dataset records sibling pairings).

Source: U.S. Social Security Administration, baby names data, births in 2025, released May 2026. Data checked 8 November 2026. ssa.gov baby names.