How we count names
This page explains where the data comes from and how each figure on the site is calculated, including every constant in the model.
1.Data sources
| Dataset | Publisher | Coverage | Used for |
|---|---|---|---|
| Baby Names from Social Security Card Applications, national data | Social Security Administration | 1880–2025; names with 5+ births per year | First-name births by name, sex and year |
| Baby Names from Social Security Card Applications, state data | Social Security Administration | 1910–2023; names with 5+ births per state and year | State-wise distribution and top names by state |
| 2010 Census surname file | U.S. Census Bureau | April 1, 2010; 162,253 surnames with 100+ people | Exact surname counts and race/Hispanic-origin shares |
| National population estimates, Vintage 2025 | U.S. Census Bureau | July 1, 2025: 341,784,857 | Scaling older counts to today's population |
| Period life tables | SSA Office of the Chief Actuary | U.S., by age | How much of each birth decade is still alive |
Two thresholds affect results. The SSA omits names with fewer than 5 births in a year for privacy, and the Census Bureau publishes only surnames held by 100 or more people. It also suppresses small race/Hispanic-origin groups, which we show as “suppressed”.
2.First names: from births to living people
SSA files count how many babies received a name, not how many of them are alive today. Adding them up directly would make names from the 1920s look as common as names from the 2010s, so we weight each decade of births by the share of that generation still living:
living(name) = Σ births(name, decade) × survival(decade) × 1.1
| Decade of birth | Age in 2025 | Share still alive |
|---|---|---|
| 1910s | 106–115 | 0.15% |
| 1920s | 96–105 | 2.0% |
| 1930s | 86–95 | 13.0% |
| 1940s | 76–85 | 38.0% |
| 1950s | 66–75 | 62.0% |
| 1960s | 56–65 | 78.0% |
| 1970s | 46–55 | 88.5% |
| 1980s | 36–45 | 95.5% |
| 1990s | 26–35 | 97.7% |
| 2000s | 16–25 | 98.6% |
| 2010s | 6–15 | 99.3% |
| 2020s | 0–5 | 99.5% |
| 1880s–1900s | 116+ | ≈0% |
Survival shares are smoothed from SSA period life tables. The factor of 1.1 corrects for under-registration in early decades and for people who got a Social Security number later in life. Across all names the model adds up to about 287,068,572 living people, against a total U.S. population of 341,784,857. The gap is mostly people born abroad, who don't appear in U.S. birth records.
3.Last names: exact counts, scaled
Last names don't need a survival model, because the Census counted living people directly. We take the exact 2010 figure and scale it by population growth since then:
today(surname) = census2010(surname) × (341,784,857 ÷ 308,745,538)
4.Full names: the independence model
No U.S. agency publishes how many people have a particular first-and-last name combination. Like other name tools, we estimate it by treating the first name and the last name as independent:
people(first, last) ≈ living(first) × census2010(last) ÷ 308,745,538
Worked example: James Smith
3,078,645 living people named James × 2,442,977 people named Smith in 2010 ÷ 308,745,538 ≈ 24,360 people
Where this bends: names cluster by community. A pairing like Jose Garcia is more common than independence predicts, because both names are common in Hispanic communities, while pairings across communities are less common. For most everyday names the approximation holds well, and every result is labeled as an estimate.
5.Scores, ages and states
Popularity score and commonness index. We sort every American by how common their name is and report the share whose name is rarer than yours. For first names the comparison uses living-bearer estimates, for last names the 2010 Census counts, and for full names the independence-model estimate of every first-and-last combination, each weighted by the number of people who hold it.
Character scorecard. Timelessness is the share of decades since 1900 in which the name had at least a quarter of its peak-decade births. Momentum compares births in the last five years with the five before. Modern feel places the peak year between 1900 and 2025. Gender balance is 100 when boys and girls are equal.
Age distribution. Births in each year are multiplied by the survival share for that birth decade (section 2) and grouped by age in 2025.
State-wise distribution. SSA state files (1910–2023) are weighted the same way, so the map reflects living people. Each state's share of the name is compared with its share of all such births; 1.0× means as common as nationally. The data records the state of birth, not the current state of residence.
Meanings. Name meanings and origins come from our editorial name database and standard etymological references. Where a meaning isn't well documented we show none.
6.Limitations
- Surname data is from 2010. It is the newest Census surname file with full detail. We scale it by official population growth.
- Very rare names are invisible. Names below the SSA threshold (5 births a year) or the Census threshold (100 people) don't appear in federal data, so we say a name is rarer than we can measure instead of guessing.
- Names, not identities. SSA records hold the name given at birth. Legal name changes, nicknames and later spellings are not modeled, and people born abroad are missing from the first-name data.
- Estimates, not official statistics. Expect figures within a few percent for common names and wider margins for rare ones.
Spotted an error? Let us know.