The 1990 piece that circulates under Spy magazine's name starts from 378 million children and assumes 3.5 children a household. It counts 91.8 million homes (378 ÷ 3.5 is 108), spreads them over the 31 hours that time zones give one night, and gets 822.6 visits a second. Every version since does the same one night. Nicholas of Myra traditionally died on 6 December 343, and tonight is the 1,684th Christmas Eve since (2026 − 343 + 1). I added up all of them.
If Santa has owed a visit to every Christian child under 15 on every Christmas Eve since 343, the total is about 106 billion visits, and the range is 85 to 145 billion. The sum multiplies three published series year by year. They are the number of Christians alive, the share of a population under 15, and one visit each per year. The range comes almost entirely from the pre-1900 Christian count, which nobody measured. For comparison, the Population Reference Bureau puts everyone ever born, of any faith, at 117 billion as of 2022.
This covers the arithmetic, the four defensible start dates, and the sources behind each input. It does not cover lumps of coal (there is no series), households that keep Sinterklaas or the Befana instead, or whether the visits happened.
When the Clock Starts
The start date is a choice, so I ran four. The first is 343, the traditional death date. Nicholas wasn't canonised in any formal sense, since no such process existed in the fourth century, so "sainthood" here means the start of the cult. The earliest Life, by Michael the Archimandrite, was written in the early ninth century, nearly 500 years later, which makes 343 a tradition rather than a record.
The second is 527, the first year of Justinian's reign. Procopius records that Justinian renovated churches in Constantinople dedicated to Nicholas and Priscus, the first dated evidence of a cult, though the buildings may go back to about 490. The third is 1087, when the relics arrived in Bari on 9 May and the Western cult took off. The fourth is 1823. "Account of a Visit from St. Nicholas" ran in the Troy Sentinel on 23 December 1823 and moved the visit to Christmas Eve, which is the night this essay is counting. A reader who holds that Santa only exists from the poem on owes the 1823 figure.
The First Council of Nicaea met in 325, eighteen years before the first start date, so no visits are owed for the council year.
Three Series, One Product
The visits in a year are the Christians alive that year times the share of them under the cutoff. Christians I took as counts, not shares, because the early sources give counts and a share would need a world-population denominator from a different source.
| Year | Christians | Source |
|---|---|---|
| 350 | 33.9 million | Stark 1996, table 1.1 |
| 1000 | 50 million | Kieffer, in TIME, 1933 |
| 1500 | 100 million | Kieffer, Europe's population |
| 1800 | 204 million | Gordon-Conwell, via a reproduction |
| 1900 | 558.3 million | Status of Global Christianity 2026 |
| 1970 | 1,222.9 million | same |
| 2000 | 1,993.4 million | same |
| 2020 | 2,526.0 million | same |
| 2026 | 2,674.0 million | same |
Stark's 350 figure is a model, not a census. He assumes 1,000 Christians in the year 40 and 40 percent growth a decade, so 350 is 1,000 × 1.4^31, or 33.9 million. He puts that at 56.5 percent of an empire of 60 million. Before 350 I ran his model back seven years, which gives 26.8 million in 343. Kieffer's 1500 figure is all of Europe's 100 million people, which counts Europe's Jews as Christians and leaves out every Christian outside Europe, from Ethiopia to Kerala. I kept it. The Jews counted in push the figure up, the Christians left out push it down, and no better published world count turned up. Between anchors I interpolated log-linearly, one row per year.
I divided those anchors by world population from HYDE 3.3 (to 1799) and Gapminder (from 1800). That puts Christians at 15.5 percent of the world in 1000 (50 of 323.5 million), 19.9 percent in 1500 (100 of 503.1 million) and 20.8 percent in 1800 (204 of 983.1 million). For the cross-check, Pew counted 2.18 billion Christians in 2010, and the interpolated ledger has 2.24 billion that year, 2.9 percent high.
The child share is the input that barely moves. From 1950 the UN's World Population Prospects 2024 gives it directly. In 1950 it is 867.7 million under 15 of 2,493.1 million, or 34.8 percent, and in 2026 it is 1,999 million of 8,301 million, or 24.1 percent. The Status of Global Christianity table lists 1,073.6 million adults over 15 out of a 1900 world population of 1,619.6 million, which leaves 33.7 percent under 15. Before 1800 I used 37.5 percent. Model life tables give 35 to 40 percent for a population with life expectancy around 25 to 35, so 37.5 is the middle and I ran 35 and 40 as the bounds. I assumed Christians have the same age structure as everyone else.
For the under-12 cutoff, the UN series gives ages 0 to 4 and 5 to 14. I took 0 to 4 plus seven tenths of 5 to 14, which gives 28.5 percent in 1950 and 19.2 percent in 2026. Before 1950 I applied the 1950 ratio of 0.82 to the under-15 share. The age at which the visits stop is a household decision, so both cutoffs are in every table.
source: content/writing-sources/outlines/101-archive/: 101-ledger.csv lines 2-1685; owid-population.csv World lines 57672-57919 (HYDE 3.3, Gapminder, UN WPP 2024); owid-population-by-age-group-with-projections.csv line 39036; gc-status-2026.txt line 18; SOURCES.md lines 8-9 (Stark 1996 table 1.1; Kieffer in TIME, 1933) · everyChild.data.ts
The Sum, Four Ways
The ledger has 1,684 rows. Each row's visits are its Christian count times its child share, and the total is the sum of the rows from the start year to 2026, tonight included.
| Start | Under 15 | Under 12 |
|---|---|---|
| 343, death of Nicholas | 106.2 billion | 86.4 billion |
| 527, Justinian's churches | 103.7 billion | 84.4 billion |
| 1087, relics to Bari | 94.3 billion | 76.7 billion |
| 1823, the poem | 64.1 billion | 52.0 billion |
The first three rows are close because the early centuries hold so few children. On the Christmas Eve of 350 Santa owed 33.9 million × 0.375, or 12.7 million visits. In 1000 it was 18.8 million, in 1500 37.5 million, in 1900 558.3 million × 0.337, or 188.2 million. Tonight it is 2,674.0 million × 0.241, about 644 million. The 184 years from 343 to 527 add only 2.4 billion to the total, and the 1,480 years from 343 to 1823 add 42.0 billion, or 40 percent. The 127 Christmas Eves from 1900 to 2026 add 53.7 billion, which is 50.6 percent of the broad total, and the 77 since 1950 add 38.6 percent.
Against the Population Reference Bureau's 117.0 billion people ever born, the broad under-15 count is 106.2 billion, or 91 percent of it. The strict count from 1823 is 64.1 billion, and the broad count is 1.66 times the strict one.
Tonight alone, spread over Spy's 31 hours of 3,600 seconds each, 644 million visits is 644,000,000 ÷ 111,600, or 5,771 a second, seven times Spy's 822.6. Spy counted homes, at 3.5 children a home. The ledger counts children, because the Christian count is a headcount and the sources give no household series that goes back to 343.
source: content/writing-sources/outlines/101-archive/101-ledger.csv lines 2-1685, columns visits_under15 and cumulative_visits_under15_from_343 (line 1685: 106,157,506,961) · everyChild.data.ts
Statistics From the Same Ledger
The same sources support other counts, and I computed them the same way.
| Quantity | Value | Source |
|---|---|---|
| Christmas Eves covered, 343 to 2026 | 1,684 | 2026 − 343 + 1 |
| Christian births, 343 to 2026 | 11.8 billion | ledger × crude birth rate |
| Visits per Christian child born | 9.0 | 106.2 ÷ 11.8 |
| Heretics slapped, First Council of Nicaea, 325 | 1 | Petrus de Natalibus, about 1370 |
| Years from the council to the first written slap | about 1,045 | 1370 − 325 |
| Slaps per century since the council | 0.06 | 1 ÷ 17.01 centuries |
| Slaps attested by anyone present | 0 | Athanasius and Eusebius never mention Nicholas |
| Bishops reinstated overnight by Christ and the Virgin | 1 | Damaskenos, printed 1570 |
| Lumps of coal | not computed | no series |
The slap row is sourced like the others. The earliest text with any striking is the Catalogus sanctorum of Petrus de Natalibus, bishop of Equilio in 1370. Nicholas strikes "a certain Arian" in the jaw and loses his mitre and pallium, and two angels give them back. The first text that names Arius as the man struck is the Life in Damaskenos's Thesaurus, printed in Venice in 1570. There the bishops jail Nicholas, Christ hands him a gospel book and the Theotokos gives him the omophorion. Nicholas's name is on Theodore Lector's attendance list from 510 to 515 and on the longer lists, and not on the shorter ones, which most scholars read as a later addition. The ninth-century Lives that put him at the council have no slap in them.
Before 1950 the births row uses the Population Reference Bureau's assumptions: 60 births per 1,000 people a year, falling to 50 by 1750 and 40 by 1850. After 1950 it uses the UN's crude birth rate, held at the 2023 value of 16.3 for 2024 to 2026. A child born in 1900 who lived to 15 got fifteen visits, and the 9.0 average is lower than that because so many children before 1900 died before they were 15.
What the Number Is Not
The total counts visits, not children. A child counted at 6 is counted again at 7, which is why 106.2 billion visits come from 11.8 billion Christian births. A reader who wants distinct children wants the births row, and should note it includes the ones who died in their first year and got no visit at all, or one.
The strongest objection is that Santa Claus as a Christmas Eve visitor is an Anglo-American figure of the nineteenth century. The broad count bills him for children whose households never expected him. Sinterklaas comes on 5 December, the Christkind on Christmas Eve in German-speaking Europe, Ded Moroz at New Year and the Befana on 6 January. That objection is right, and the 1823 row is the answer to it, but only half of one. Even from 1823 the ledger counts every Christian child on earth, not the households that kept the custom, and I found no series that splits them. I report both totals: 106.2 billion for the premise as stated, and 64.1 billion for a reader who holds Santa to the poem.
Limits
The pre-1900 Christian count is the weakest input. Stark's 350 is a growth model. Kieffer's 1000 and 1500 are a 1933 statistician's round numbers. The 1800 figure is a secondary reproduction of a Gordon-Conwell table I couldn't find in the original. Halving the 1000, 1500 and 1800 anchors gives 85.2 billion from 343, and doubling them gives 145.5 billion. That is the range in the second paragraph, and I can't state the answer more tightly than that. Halving or doubling Stark's 350 with them gives 83.1 or 150.9 billion.
The other inputs move it less. The pre-1800 child share at 35 or 40 percent instead of 37.5 gives 103.1 or 109.2 billion. Linear interpolation between anchors instead of log-linear gives 110.0 billion from 343 and 66.5 billion from 1823, 3.6 and 3.7 percent above the log-linear totals. The age structure before 1950 is a model, not a census, and the Spy figures are quoted from the copies that circulate, not re-derived.
source: content/writing-sources/outlines/101-archive/101-ledger.csv lines 2-1685, visits_under15 and visits_under12 summed from each start year; sensitivity runs, drafts/101-every-child-since-myra-factcheck.md recomputation table; PRB, Kaneda and Haub 2022 (SOURCES.md line 11) · everyChild.data.ts
The Count Since 1823
The broad count is 106.2 billion and the strict count is 64.1 billion. The 42.0 billion between them (106.16 − 64.13 before rounding) are the visits owed from 343 to 1822, the 1,480 Christmas Eves before the poem put Santa on that night. Every input and every interpolated year is in the published ledger, one row per Christmas Eve. A reader with a better count of the Christians alive in the year 1000 can replace one cell and rerun the sum.