New Numerology API

Measured facts

Measured facts about the Astrology and Numerology APIs

API for Astrology (apiforastrology.com) measures its astrology and numerology APIs and publishes the results. 75.0% of 10,000 random birth charts are manglik (the Mars check in marriage matching), before exceptions. 16.3% of birth dates from 1926 to 2025 have a master-number bhagyank (destiny number). All 161 festival dates checked matched a published Hindu calendar.
See the accuracy testsRead the rulebook

Updated

Key numbers

  • 75.0%

    of 10,000 random birth charts are manglik before exceptions, from the rising sign or the Moon

  • 16.3%

    of 36,525 birth dates (1926 to 2025) have a master-number bhagyank

  • 161 of 161

    festival dates matched a published calendar

  • 480

    written rules, each with its own page

How close are the Astrology API's planet positions to NASA's?

One arc second (″) is 1/3,600 of a degree. At the same moment, the Sun and planets match NASA JPL Horizons within 0.001″. The Moon matches within 0.011″, over 25,571 positions from 1850 to 2149. From a birth clock time, on 2,000 random charts, the Sun and planets came within 0.06″ of Swiss Ephemeris. The Moon came within 0.41″.

FactNumberHow it was measuredSampleMeasuredUsed on
Sun and planet positions against NASA JPL Horizons, at the same momentWithin 0.001″21,918 positionsWhere the Sun, Mercury, Venus, Mars, Jupiter and Saturn are, seen from Earth's center, compared with NASA JPL Horizons at the same moment. The check runs as a test in the API's code.Every 30 days from 1850 to 2149Accuracy tests
Moon positions against NASA JPL Horizons, at the same momentWithin 0.011″3,653 positions, median 0.0025″The same check for the Moon. The gap grows toward both ends of the range, where two editions of NASA's Moon data differ slightly.Every 30 days from 1850 to 2149Accuracy tests
Sun and planet positions from a birth clock timeWithin 0.06″Random birth moments and places, cast from their clock time and compared with Swiss Ephemeris.2,000 random charts from 1850 to 2026Accuracy tests
Moon positions from a birth clock timeWithin 0.41″Median 0.013″The same 2,000 charts. The Moon's gap comes from delta T, the small, changing gap between atomic clock time and Earth's spin. Each program reads delta T from its own table.2,000 random charts from 1850 to 2026Accuracy tests
Rahu and Ketu (the true node) from a birth clock timeWithin 0.12″Rahu and Ketu are the two points where the Moon's path crosses the Sun's path. The same 2,000 charts, compared with Swiss Ephemeris.2,000 random charts from 1850 to 2026Accuracy tests
Rising sign from a birth clock timeWithin 0.004″The same 2,000 charts, at places from 60° south to 60° north, compared with Swiss Ephemeris.2,000 random charts from 1850 to 2026Accuracy tests
Lahiri ayanamsa (the zodiac offset)Within 0.001″Largest gap 0.0006″Compared with Swiss Ephemeris in the API's automated tests.8 dates from 1900 to 2100Accuracy tests

Each zodiac sign is 30 degrees wide, which is 108,000 arc seconds.

Sources: NASA JPL Horizons, every position compared (JSON download)

How common is manglik dosha in birth charts?

In 10,000 random birth charts, 75.0% were manglik before exceptions. Manglik dosha is a Mars check in marriage matching. The API counts the house of Mars from the ascendant (rising sign) and the Moon. Another 13.6% were partial, found only from Venus. In 33.8%, a check found the dosha and the API named an exception that can cancel it.

FactNumberHow it was measuredSampleMeasuredUsed on
Birth charts that are manglik, found from the ascendant or the Moon, before exceptions75.0%7,502 of 10,000 chartsMars in house 1, 2, 4, 7, 8 or 12, counted from the ascendant and from the Moon. A chart counts when either finds it, as the API's manglik verdict does. Run with the API's own manglik code.10,000 random birth charts: births from 1940 to 2010, places from 40°S to 60°N
Birth charts that are partly manglik, found from Venus only13.6%1,358 of 10,000 chartsThe same check counted from Venus finds it, but the ascendant and the Moon do not. The API marks these charts partial, not manglik.10,000 random birth charts: births from 1940 to 2010, places from 40°S to 60°N
Birth charts that are manglik from the ascendant alone50.5%5,055 of 10,000 chartsThe same check, with the house of Mars counted only from the ascendant (rising sign).10,000 random birth charts: births from 1940 to 2010, places from 40°S to 60°N
Birth charts found manglik by any check that also have a named exception33.8%3,379 of 10,000 chartsA check from the ascendant, the Moon or Venus finds Mars, and the API names an exception. One example is Mars in its own sign. The API reports exceptions but does not apply them. Weighing them is up to the astrologer.10,000 random birth charts: births from 1940 to 2010, places from 40°S to 60°N

How often do master numbers show up in birth dates?

Of every date from 1926 to 2025, 16.3% have a master-number bhagyank (destiny number). Its digit total is 11, 22 or 33, or passes through one of them. Counted the Western way, the share is higher. On 44.56% of the same dates, a Western life path, birthday or pinnacle (life stage) is a master number.

FactNumberHow it was measuredSampleMeasuredUsed on
Birth dates with a master-number bhagyank16.3%5,959 of 36,525 datesThe digits of the day, month and year are added. The number is a master number when the total is 11, 22 or 33. It also counts if the total passes one of them on the way down. Run with the API's own bhagyank code (POST /v1/profile).Every date from Jan 1, 1926 to Dec 31, 2025: 36,525 datesHome page
Birth dates with a Western master number, from the date alone44.56%16,274 of 36,525 datesA master number (11, 22 or 33) in the life path, the birthday number or any of the four pinnacles. Run with the API's own Western code (POST /v1/western). We also sent 150 random dates to the API, and all 150 matched.Every date from Jan 1, 1926 to Dec 31, 2025: 36,525 datesMaster numbers post

The two shares count different numbers, so both are right. The bhagyank share looks at one Indian number. The Western share looks at six: the life path, the birthday number and four pinnacles.

How well do festival, fasting and muhurat dates match a published calendar?

All 161 festival dates from the Astrology API matched a widely used published calendar. Those dates cover New Delhi from 2023 to 2028 and Beijing in 2026. Of 496 fasting days (vrats), 495 matched. All 579 published muhurat windows (good times to start something) matched within 3 minutes at both ends.

FactNumberHow it was measuredSampleMeasuredUsed on
Hindu festival dates that match the published calendar161 of 16123 festivals, such as Diwali and Holi, worked out by the API. Each was compared with the published calendar's pages, read Sep 16, 2026.New Delhi from 2023 to 2028 and Beijing in 2026: 7 calendarsAccuracy tests
Fasting days (vrats) that match the published calendar495 of 496Ekadashi (the 11th lunar day), Pradosh, Sankashti, Masik Shivaratri, Purnima (full moon) and Amavasya (new moon). Each was compared with the published calendar's pages, read Sep 25, 2026. The one miss is a full moon that ends 57 seconds after sunrise in New Delhi on Mar 11, 2028. The API counts that sunrise, and the calendar does not.New Delhi in 2026, 2027 and 2028, and Beijing in 2026Accuracy tests
Muhurat windows for buying a vehicle or property and moving into a home418 of 418The start and end of each window, compared with the published calendar's New Delhi pages, read Sep 16, 2026. A match is within 3 minutes at both ends.New Delhi, 2026 and 2027: 200 vehicle, 136 property and 82 griha pravesh (moving in) windowsAccuracy tests
Marriage (vivah) muhurat windows161 of 161The same check, within 3 minutes at both ends. The API also gives two short 2027 windows that the calendar does not list.New Delhi, 2026 and 2027Accuracy tests

Each check stays in the API's code as a test, so a later change cannot quietly undo a matched date.

How many rules, endpoints and glossary terms do the APIs have?

The Astrology API follows 480 written rules in 42 sections, and each rule has its own page. In 284 of them the traditions differ, so the rule states the choice the API makes. The Astrology API has 65 endpoints and the Numerology API has 16. The glossaries explain 216 astrology terms and 47 numerology terms.

FactNumberHow it was measuredSampleMeasuredUsed on
Written rules the Astrology API follows480Counted from the API's own rulebook, in 42 sections.Every ruleLive countRulebook
Rules where the traditions differ and the API states its choice284Counted from the API's own rulebook, where each rule is marked with its kind.Every ruleLive countRulebook
Rules from the classical texts, where the major traditions agree123Counted from the API's own rulebook, where each rule is marked with its kind.Every ruleLive countRulebook
Rules the API adds where the texts give no method59Counted from the API's own rulebook, where each rule is marked with its kind.Every ruleLive countRulebook
Rules that name something the API does not compute, so no one assumes it does14Counted from the API's own rulebook, where each rule is marked with its kind.Every ruleLive countRulebook
Astrology API endpoints65Counted from the public API reference. Only /v1 routes count.Every public routeLive countAstrology API reference
Numerology API endpoints16Counted from the public API reference. Only /v1 routes count.Every public routeLive countNumerology API reference
Astrology glossary terms216Counted from the glossary, one page per term.Every termLive countAstrology glossary
Numerology glossary terms47Counted from the glossary, one page per term.Every termLive countNumerology glossary

These counts are read from the API's data each time the site is built, so they stay current.

Astrology API

Run your own checks

Create a free test key and send your own charts and dates. Or try a request in the API reference first.

Get a free API keyOpen the API reference

Your first answer

John Doe, New York

POST /v1/planets200 OK
  • RisingVirgoHast
  • SunCancerAshlesha
  • MoonTaurusMrigashira
  • MarsAriesKrittika

Real API output. A free test key gets the same answer.

  1. 01

    Test key: 1,000 test calls a month to build with

  2. 02

    Free plan: 2,000 live calls a month, no card needed

  3. 03

    Open rules: every rule is published, so you can check any answer