The Answer in One Number
Follow every Top-100 name from its child's birth year to that child's 18th birthday, across thirty birth cohorts (1978–2007): 38% of girls' names and 56% of boys' names are still in the Top 100 at graduation. The typical girl's name has drifted to about #144; the typical boy's name to #82. Boys' names age slowly; girls' names are fashion.
Every naming decision contains a hidden bet about the future: the name you choose from today's landscape will be worn in a landscape eighteen years away. Most popularity advice ignores this entirely — it tells you where a name ranks now, as if your child will be a baby forever. So we measured the thing parents actually want to know. Take a name from the Top 100 in a birth year. Fast-forward eighteen years. Where is it? Done once, that's an anecdote; done for every Top-100 name across thirty birth years — 6,002 name-cohorts checked against the full official record — it's a shelf life.
The single most useful finding is the gap between the sexes. A boy's Top-100 name is a reasonably stable asset: more often than not, it's still a Top-100 name when he's grown, and it almost never falls far — across all thirty cohorts, only 13 of 3,000 boys' name-cohorts (0.4%) had fallen below #500 by age 18. A girl's Top-100 name is a different instrument: 317 of 3,002 (10.6%) fell below #500 over the same horizon. Neither is better. But they are different products with different warranties, and almost no naming advice tells you so.
How We Measured It, and What It Does Not Mean
The measure is deliberately simple. For each birth year from 1978 through 2007, take the names in that year's Top 100 (girls and boys separately). Then look up each name's rank eighteen years later. "Survived" means: still in the Top 100. That's it — no adjustments, no weighting, nothing tunable.
One sentence of honesty before you use it: this measures where the name sits on the newborn charts eighteen years later — not how common the name is among your child's own classmates. A name your child shares with five kindergarten classmates will still be everywhere in their grade at eighteen, whatever the newborn lists say by then. Shelf life tells you whether the name will feel current or of its era to the next generation of parents — the people naming your child's younger cousins. That's a real thing to care about; it just isn't the same thing as crowding, which is set at birth.
The 2007 Cohort at Age 18, Measured Against 2025 Births
The newest cohort with a complete window: children born in 2007 turned eighteen into the 2025 lists. Of their era's hundred most-popular girls' names, 37 are still Top-100 names today; of the boys' hundred, 54. Right on the thirty-year averages — this is a stable property of the naming system, not a quirk of any one generation.
We keep the full 200-name outcome table out of the article on purpose — it's a lookup, not a story. The complete cohort list ships as a downloadable CSV alongside this page, and if you want to know where any specific name sits today, look it up in the name tool — it carries the same official data this study is built on.
Shelf Life Collapsed for a Century, Then Stopped Falling
Is 38% good or bad? It depends when you ask. Run the same measurement back to the very beginning of the record and the long arc appears:
| Birth decade | Girls' survival | Boys' survival | Birth decade | Girls' | Boys' |
|---|---|---|---|---|---|
| 1880s | 77.0% | 82.4% | 1950s | 45.6% | 62.0% |
| 1890s | 76.7% | 82.2% | 1960s | 41.7% | 55.5% |
| 1900s | 72.6% | 84.4% | |||
| 1970s | 46.2% ↑ | 63.8% ↑ | 1910s | 61.9% | 79.1% |
| 1920s | 56.0% | 72.5% | 1980s | 42.1% | 57.3% |
| 1930s | 54.7% | 70.8% | 1990s | 33.0% | 55.6% |
| 1940s | 53.0% | 69.9% | |||
| 2000s | 36.4% ↑ | 54.8% |
Three readings, in order of importance:
- The century-long collapse is real but finished. A girl born in the 1880s could expect 77% of her era's popular names to greet her adulthood; by the 1990s, 33%. But the fall was not a straight line to zero — it flattened out two generations ago.
- The 1970s uptick is the quiet surprise — survival rose for both sexes (girls 41.7%→46.2%, boys 55.5%→63.8%) as a wave of durable staples settled in. And the girls' series ticked up again from the 1990s (33.0%) to the 2000s (36.4%). Two upticks aren't destiny, but they say today's parents are choosing from the most stable naming era in half a century — not a landscape in free fall.
- The sexes never converge. In no decade, ever, did girls' names outlast boys'. The gap is a permanent feature of American naming, visible from the first cohort to the last.
The Scale-Free Control: Share Retention by Peak Era
A fair objection: the Top 100 itself changed underneath this measurement. In 1900 the girls' file held 2,223 rankable names and the Top 100 covered 68.0% of all girls' births; by 2007 the pool was 20,584 names and the Top 100 covered just 35.7%. Rank-based shelf life is not a scale-free quantity — part of the early decades' near-invincibility is simply that the list was shallow. So here is the same question asked in a way list depth can't distort: when a name hits its all-time peak share of births, how much of that share does it keep twenty years later?
Same direction, honest scale: names really do fade faster than they did a century ago, even after removing the list-depth artifact. The background force is the fragmentation of naming itself — in 1880 the ten most popular girls' names covered 24.65% of all girls born; in 2025, 6.99% (boys: 44.22% → 7.98%). Every name now holds a smaller slice of a wider table, so every fall is from a lower height. (That concentration series lives on this page; our long-tail study covers the width of the table itself.)
Five Names That Show Each Pattern
Numbers only — each name here is an exhibit for one trajectory, not a recommendation:
What should a parent actually do with all this? Not chase stability — a name is not an index fund. But the measurement does convert one worry into a fact: if you love a Top-100 girls' name, expect its chart position to age visibly by graduation (median landing: #144, entirely alive and well); if you love a Top-100 boys' name, expect it to still be roughly where you left it. Run your own shortlist against the current lists in the generator, decide with open eyes, and then — per the naming timeline — check the data exactly once and get back to the parts of naming that are actually about your child. If what you want is a name that has already survived every era of this chart, that's a different, curated question — Names That Age Well answers it (this page measures; that page picks).
Limits, Thresholds, and Where the Numbers Bend
- Source: official SSA national name files, 2025 release (published May 8, 2026). All computation by Momcozy; no third-party name data.
- Cohort definition: a cohort is a birth year's Top-100 names per sex; observation year = birth year + 18. Headline pool: birth years 1978–2007 (n = 3,002 girls' / 3,000 boys' name-cohorts — ties occasionally put 101 names in a "Top 100"). Exact pooled survival: girls 1,140/3,002 = 37.97%; boys 1,686/3,000 = 56.20% — rounded to 38%/56% in the headline.
- "Fell below #500" means published rank > 500 or absent from the published file in the observation year: girls 317/3,002 (10.6%), boys 13/3,000 (0.4%).
- Medians are pooled across all name-cohorts (girls #144.0, boys #82.0). Cohort-by-cohort medians differ slightly; we use one convention throughout and name it here.
- Ties share a rank (SSA convention): in 2025, girls' #24 is held by two names at 6,176 births each, and #25 is skipped. Rank arithmetic in this article inherits that convention.
- No right-censoring in the headline: the newest cohort (2007) has a complete 18-year window; nothing after birth-year 2007 enters the survival figures. In the peak-retention control, names still rising in 2025 have undefined peaks and are excluded by the peak-year cutoffs (1929 / 1989).
- Cross-era comparability is limited by list depth — the rankable pool grew from 2,223 girls' names (1900) to 20,584 (2007), and Top-100 coverage of girls' births fell 68.0% → 35.7%. That is why section 5's scale-free control exists; read the decade chart with that caveat attached.
Keep going: How to choose a baby name, step by step · Naming your second baby · Announcement ideas
Original analysis by Momcozy from official SSA files under the definitions in the methodology box; the full cohort-outcome table is available as a downloadable CSV. Names appear as numeric exhibits only — no meanings, origins, or profiles. Nothing here ranks names by quality; shelf life is one property among many, and not the most important one.
