Methodology and sources
Every Boy Name shows only numbers that come from official birth registrations, published by the statistics office or registry of each country, province or city. Nothing is estimated from surveys, forum votes or search trends.
Sources, by region
42 official sources
Ranks and counts
A rank is the name's place among boys in that publisher's list for that year. Where the publisher gives counts, tied names share a rank (1, 2, 2, 4). Where it gives ranks only, we show its ranks.
Spellings that differ only by accents or spaces are counted as one name (Théo with Theo). The rank shown for a region refers to the spelling published there.
Publishers leave out their rarest names, usually those given fewer than 3 or 5 times, and France rounds every count to a multiple of 5. Munich counts children aged under one living in the city at the end of the year rather than births.
Shares and “1 in N”
Where a publisher lists nearly every name given (a full list), we divide by all the boys in that list to get a share per 100,000 boys, and “1 in N” is the same share turned round. Top-N lists cannot give a share, so we compare them by rank only.
Loved, common and rare
Each place is drawn in one of three tones, from the name's rank in that place's latest official list: loved is number 1 to 10, common is 11 to 100, and rare is 101 or below, or missing from a full list (one that names nearly every boy). A place that publishes only a top list without the name has no figures for it and is left blank.
The sentence above each strip and map names the places in each tone, the biggest lists first. Very short lists, and lists too small for a page of their own, are named only when nothing else shares their tone.
Some shares run high: a list that covers a single city, a publisher that drops names below a cut-off of five babies or more, or a list that misses more than a tenth of the births. Those shares say “about” and never make the headline “1 in N”.
Trends, rising names and generations
A trend compares the latest year with five years earlier. A name is rising in a region when its share went up by a quarter or more, or when it climbed 10 or more places in a top list, and falling on the reverse.
Generations come from the longest records: France since 1900, and Ontario and British Columbia combined. The median birth year is the year by which half of all the boys ever given the name there had been born.
Styles such as vintage, revival and classic are defined by rules over these records, and each style page prints its rule.
Name fit
Two names count as the same style when their popularity rose and fell together across every region and decade we hold. On top of that, the engine compares median birth years, popularity and origin, and checks sound: rhymes, a last sound that runs into the next name, shared first letters and endings, syllable rhythm and initials that spell a word.
The same engine builds the static brother, middle and surname pages and runs in your browser in the Name Finder, so the two always agree. Every point it awards is printed as a reason.
Meanings, origins and pronunciation
Meanings and origins come from Wiktionary (text under CC BY-SA 4.0) and Wikidata (CC0). We print a meaning only when a dictionary gives one, keep any “possibly” or “probably”, and show when a meaning is inherited from the name a short form comes from. Where the only record is the language a name is used in, we say “used in” rather than claiming an origin.
Syllable counts and sound checks use the CMU Pronouncing Dictionary (BSD licence). For the few names it does not list, the count is read from the spelling, and the name's page says so. The surname pages cover the most common surnames in the US Census Bureau's 2000 surname list (public domain).
What is not here
Germany publishes no official national first-name statistics, so German figures come from the registers of individual cities and are labelled as cities. Where a country publishes only a short top list, a name outside it is simply not shown for that country.