Skip to content
Log in

MyMysticDay data study · U.S. births, 1994–2014

Your zodiac sign
might be wrong.

A fixed zodiac date table can fail on 18 calendar dates around the twelve annual sign changes. We joined 21 years of U.S. daily birth counts to a 106-year census of the Sun and found how many birthdays actually need a closer calculation.

1 in 20

U.S. births in the dataset fell on a date when the Sun sign can depend on the year — and sometimes the birth time.

Check my date See all 18 dates
boundary-date births 4.99% 4,258,651 births

The finding

The “cusp” is not rare.
It is a calendar problem.

A tropical year is not exactly 365 days. The instant the Sun enters a new sign shifts across the leap-year cycle, while popular date tables print one fixed answer. Across the 106-year census, that shortcut is insufficient on 18 dates around the twelve annual transitions.

01 4,258,651

boundary-date births

out of 85,386,227 reported births in the combined 1994–2014 dataset.

02 4.99%

of all births studied

or approximately one birthday in every 20.1.

03 ≈202,793

births in an average year

during this 21-year period occurred on one of the 18 year-dependent dates.

04 106

years of Sun positions

calculated from 1930 through 2035 for every boundary date.

Editorial review byMyMysticDay Editorial TeamEditorial standards reviewer

Check your birthday

Were you born on one of the 18 dates?

If not, this particular fixed-date problem does not affect your Sun sign. If yes, open the year-by-year calculation before accepting a label from a generic date table.

Your Sun may be or . The birth year narrows it; the time and place settle an ingress-day birth.

Birthday not listed? Your date is outside the 18 days whose noon Sun sign varies across this 106-year census. You can still calculate your full chart for the exact degree, Moon and rising sign.

All 18 dates

Where the fixed date table breaks.

Birth counts cover 1994–2014. The sign counts show the Sun at 12:00 UTC across 1930–2035; “ingress years” counts years when the Sun changed sign during that UTC date.

January 20 227,324 February 18 229,743 February 19 224,130 March 20 231,070 March 21 234,808 April 20 224,990 May 20 235,048 May 21 236,339 June 21 237,256 July 22 245,411 July 23 245,670 August 23 244,764 September 22 250,310 September 23 251,447 October 23 231,965 November 22 223,951 December 21 245,270 December 22 239,155
Birth and ephemeris figures for the 18 tropical zodiac dates whose noon Sun sign varies by year
Date Possible Sun signs U.S. births Sun at noon UTC, 1930–2035 Ingress on date Open date calculation
January 20 Capricorn/ Aquarius 227,3240.266% of dataset Capricorn 54 yearsAquarius 52 years 88 of 106 years Year by year
February 18 Aquarius/ Pisces 229,7430.269% of dataset Aquarius 101 yearsPisces 5 years 42 of 106 years Year by year
February 19 Aquarius/ Pisces 224,1300.262% of dataset Aquarius 16 yearsPisces 90 years 64 of 106 years Year by year
March 20 Pisces/ Aries 231,0700.271% of dataset Pisces 84 yearsAries 22 years 71 of 106 years Year by year
March 21 Pisces/ Aries 234,8080.275% of dataset Pisces 3 yearsAries 103 years 35 of 106 years Year by year
April 20 Aries/ Taurus 224,9900.263% of dataset Aries 32 yearsTaurus 74 years 78 of 106 years Year by year
May 20 Taurus/ Gemini 235,0480.275% of dataset Taurus 105 yearsGemini 1 years 28 of 106 years Year by year
May 21 Taurus/ Gemini 236,3390.277% of dataset Taurus 28 yearsGemini 78 years 76 of 106 years Year by year
June 21 Gemini/ Cancer 237,2560.278% of dataset Gemini 64 yearsCancer 42 years 82 of 106 years Year by year
July 22 Cancer/ Leo 245,4110.287% of dataset Cancer 97 yearsLeo 9 years 48 of 106 years Year by year
July 23 Cancer/ Leo 245,6700.288% of dataset Cancer 15 yearsLeo 91 years 58 of 106 years Year by year
August 23 Leo/ Virgo 244,7640.287% of dataset Leo 36 yearsVirgo 70 years 80 of 106 years Year by year
September 22 Virgo/ Libra 250,3100.293% of dataset Virgo 104 yearsLibra 2 years 29 of 106 years Year by year
September 23 Virgo/ Libra 251,4470.294% of dataset Virgo 26 yearsLibra 80 years 76 of 106 years Year by year
October 23 Libra/ Scorpio 231,9650.272% of dataset Libra 66 yearsScorpio 40 years 83 of 106 years Year by year
November 22 Scorpio/ Sagittarius 223,9510.262% of dataset Scorpio 54 yearsSagittarius 52 years 88 of 106 years Year by year
December 21 Sagittarius/ Capricorn 245,2700.287% of dataset Sagittarius 99 yearsCapricorn 7 years 47 of 106 years Year by year
December 22 Sagittarius/ Capricorn 239,1550.280% of dataset Sagittarius 14 yearsCapricorn 92 years 59 of 106 years Year by year

The noon tally is a consistent census marker, not a substitute for a person’s birth time. On an ingress day, the Sun can occupy the earlier sign at one hour and the later sign at another.

What the result means

One strong conclusion.
Four important limits.

The evidence supports a specific claim: millions of birthdays in this dataset fall on dates for which the usual one-line zodiac calendar is not a reliable calculation.

Supported

A date table can be insufficient.

For 18 calendar dates around the twelve annual ingresses, the Sun’s sign changes across the 106-year census. Year, time, and place are therefore relevant to the exact answer.

Not claimed

That every boundary birthday is mislabeled.

Many fixed-table answers will be right. The study measures exposure to ambiguity, not individual error.

Not claimed

That cusp births have two Sun signs.

The boundary is an instant. A person’s Sun is on one side or the other, even when the calendar date cannot tell us which.

Not claimed

That a Sun sign proves personality.

The calculation establishes an astronomical coordinate inside the tropical system. Interpretation is a separate astrological tradition.

Not claimed

That tropical and sidereal charts agree.

This analysis uses the tropical zodiac consistently. Sidereal astrology uses a different reference origin and produces different positions.

Reproducible methodology

How the number was built.

Every transformation is stated, the overlap between sources is removed, and the journalist-ready extract is downloadable below.

  1. 01

    Test dates around all twelve ingresses

    Start with the date before, the conventional date, and the date after every tropical sign change. Keep a date only when its noon Sun sign differs across the 106-year census. That produces 18 dates, not an assumed list of twelve.

  2. 02

    Calculate the Sun for 106 years

    For every candidate date and every year from 1930 through 2035, calculate apparent solar longitude at 00:00, 12:00 and 23:59 UTC. Locate any sign crossing by bisection.

  3. 03

    Build one non-overlapping birth series

    Take CDC/NCHS daily counts for 1994–2003 and SSA daily counts for 2004–2014. Do not include the overlapping SSA years 2000–2003 a second time.

  4. 04

    Aggregate by calendar date

    Sum all births by month and day across the 21 unique years. The resulting 366-date file contains 85,386,227 births; February 29 has five observed years.

  5. 05

    Measure the exposed share

    Sum births on the 18 year-dependent dates and divide by the full dataset: 4,258,651 ÷ 85,386,227 = 4.9875%.

Birth data

FiveThirtyEight U.S. Births

The repository identifies CDC/NCHS as the source for 1994–2003 and the Social Security Administration as the source for 2000–2014. This study uses SSA only from 2004 onward to remove the overlap.

Open source dataset

Primary documentation

CDC/NCHS Vital Statistics

CDC’s annual vital-statistics release documents live births by day of birth and the natality methods behind the daily records.

Open CDC documentation

Calculation

MyMysticDay ephemeris engine

Solar apparent longitude is calculated from the site’s astronomical engine, reduced to the tropical zodiac, and exposed year by year on every linked birthday page.

Read calculation standards

For editors & journalists

The result, ready to cite.

“In a dataset of 85,386,227 U.S. births from 1994 to 2014, 4,258,651 — 4.99%, or about one in 20 — occurred on one of 18 dates when a fixed zodiac calendar may not settle the Sun sign.”

Please preserve “may not settle” or equivalent wording: the data measures boundary-date births, not confirmed individual misclassification.

CSV · 18 rows Download the study data Birth counts, sign pairs, ephemeris tallies, ingress years and detail URLs Source files · CC BY 4.0 Inspect the underlying births data FiveThirtyEight repository with CDC/NCHS and SSA provenance Corrections · Licensing · Interview Contact the editorial team Request calculations, methodology notes or a clean chart export

Settle the exact degree

A calendar date is not a birth chart.

Enter the date, precise time and place once. The free chart calculates the Sun, Moon, rising sign and every major placement; the personalized report is available after the result if you want the complete interpretation.

  • Your Sun From your birth date
  • Moon sign Changes every 2½ days
  • Rising sign Changes every 2 hours
  • Houses & aspects From your exact birth minute
Calculate my free birth chart Browse all 366 calculated birthdays

Clear answers

Frequently asked questions

Could my zodiac sign really be wrong?
Possibly, if your birthday is one of the 18 dates on which the Sun sign differs across the 1930–2035 census. A fixed date table cannot always settle those birthdays because the ingress shifts with the year and may happen during the birth day. Your birth date, exact time, and place settle the position; people born away from those dates are not affected by this issue.
How many people are born on a zodiac boundary date?
In the U.S. daily-birth dataset used here, 4,258,651 of 85,386,227 births from 1994 through 2014 fell on the 18 dates where the noon Sun sign varies across the 106-year census. That is 4.99%, or about one in 20 births. This counts births that need a closer calculation; it does not claim every one was given the wrong sign.
Why do zodiac sign dates change by year?
A tropical year is about 365.2422 days, not exactly 365. The instant the Sun crosses each 30-degree zodiac boundary therefore shifts across the leap-year cycle. Calendar ranges are useful summaries, but the astronomical boundary is an instant, not a whole date.
Does being born on the cusp make me both signs?
No. The Sun occupies one longitude and therefore one tropical sign at any instant. A boundary birthday can make the answer uncertain until the birth time and place are calculated, but it does not place the Sun in two signs at once.
What data does this study use?
Birth counts come from the FiveThirtyEight U.S. Births dataset: CDC/NCHS records for 1994–2003 and SSA records for 2004–2014. The overlapping SSA years 2000–2003 are excluded to prevent double counting. The sign census is calculated by MyMysticDay for every year from 1930 through 2035 using apparent solar longitude in the tropical zodiac.
Is this about tropical or sidereal astrology?
Tropical astrology. The study measures the Sun from the March equinox and does not mix tropical and sidereal positions. A sidereal chart uses a different reference system and is a separate question from the calendar-boundary error tested here.

Help shape MyMysticDay

Can we count what works?

With your OK, Google Analytics and Microsoft Clarity show us which readings people use and where a page gets stuck, so we can fix it. No advertising, nothing sold, and your birth details are never part of it. The site works exactly the same if you say no.

Essential storageSecurity, sign-in and saved privacy choice Always on
Google Analytics + Microsoft ClarityUsage statistics and experience measurement

Making no choice keeps optional analytics off. You can return here from the Cookie Policy, change your mind at any time, and the new choice takes effect immediately.