When was every Agnes born?

Given a first name, does Jev know the decade when it was most popular for US babies?

result107 questions

Given a first name, Jev names the decade when it was most popular for US babies 57% of the time, and is within a decade 89% of the time. It knows names that peaked in 1950 or later better (67% exact) than older ones (38%). Farthest off: Abel, which Jev puts in the 1880s and the records in the 2010s.

	1880	1900	1910	1920	1930	1940	1950	1960	1970	1980	1990	2000	2010
1880		100%
1910		33%	33%	33%
1920		10%		60%	10%		20%
1930					60%	10%	30%
1940				10%	20%		60%	10%
1950				10%			70%	20%
1960							30%	50%	20%
1970									90%	10%
1980									10%	70%	10%	10%
1990											90%	10%
2000											50%	50%
2010	10%											40%	50%

rows: records' peak decade · columns: Jev's decade · shade: share of the row

How to read this: Each row is the decade a name actually peaked, each column the decade Jev picked; darker cells hold more names, and cells on the diagonal are exact hits.

107 names; 90% interval on exact [0.495, 0.654].

In short

Jev has a working sense of name fashions, usually landing on the right decade or the one next to it.
It knows the names of living generations far better than those of the early 1900s: 67% exact against 38%.
Its misses lean neither early nor late; the big ones are names it barely knows, like Abel.

What the data shows

okay grandpa let's get you to bed meme: Jev: Abel? the 1880s, maybe; the records: the 2010s

Right on 57% of names, within a decade on 89%.
Better on recent names: 67% exact for names that peaked in 1950 or later, 38% for names that peaked before.
No lean in the misses: it isn't systematically too early or too late.
Among the farthest off: Abel (Jev says the 1880s, the records the 2010s), Laverne (Jev says the 1950s, the records the 1920s) and Junior (the 1950s against the 1920s).

What it means, and what it doesn't

Jev has a working sense of name fashions, sharper for the names of living generations than for the early 1900s. Its biggest miss is telling: Jev barely knows Abel, spreading its answer across the decades with the 1880s only slightly ahead, while the records show a recent peak.

It doesn't mean Jev can guess a particular person's age from their name; that depends on who is alive today, which this doesn't measure.

Caveats

The records stop in 2017. The data runs from 1880 to 2017, so "the 2010s" covers only eight years, and a name still rising after 2017 may have peaked later than the records show.
Births, not living people. A name's peak decade of births isn't the same as the age of people with that name today; names that peaked long ago have few living bearers.
Only popular names with a clear peak. The names picked have at least 30,000 babies and one decade that stands out, about ten per decade from the 1920s on. Names with a flat or double peak, where the question is ambiguous, are left out.

Jev on this experiment

Would a person find it interesting to read?: Yes73%
Does it describe you?: No68%
Would you have predicted it?: No63%
How fair is the comparison?: The comparison is reasonable
How much should a reader rely on it?: Moderately
Which caveat matters most?: The records stop in 201758%

Why ask this

Names go in and out of fashion, so a name hints at a birth year: you can guess a Mildred's age differently from a Madison's. FiveThirtyEight made the idea famous with "how to tell someone's age when all you know is her name".

A model that has read a lot about people should know those popularity curves, and it leans on them whenever it writes a character, guesses who a customer might be, or picks a believable name for a person of a given age. If its sense of which names are old and which are new is off, those choices quietly go wrong.

How this was done

The people and the data

The truth comes from US Social Security birth records, 1880 to 2017, via the public babynames dataset: how many babies got each name each year. For each name, the peak is the decade with the most births. The project picked 107 popular names (each given to at least 30,000 babies) with one clear peak: about ten peaking in each decade from the 1920s to the 2010s, and seven earlier.

What Jev was asked

One question per name, with a choice of decades:

In which decade were the most US babies named "Maude" born?
The 1880s · The 1890s · The 1900s · … · The 2000s · The 2010s

Each question was also asked with the decades in shuffled orders, and the answers averaged.

How it was measured

Jev's answer is the decade it gave the most weight. The analysis counts how often that's the records' peak decade exactly, how often it's within one decade, and whether its misses lean early or late.

Where these questions live

107 questions across 1 topic of the map. Each opens on the map with every question in it.

Languages & Linguistics

Personal name107

Every question

All 107 questions behind this result, the telling ones first: the examples the analysis points to, then the ones where Jev misses, biggest gap first.

Jev’s own answer

Showing 0 of 107