2.33° apart · measured on 10,000 births

Why two apps give you a different Moon sign

The ayanamsa systems disagree by 2.33 degrees. Measured on 10,000 births: different Moon rashi 4.6% of the time, different nakshatra pada 42.6%, and dasha dates move by months.

Because they are measuring from different starting points, and the starting points disagree by 2.33 degrees — about 140 arc-minutes. Neither app is broken. They have made different choices about where the zodiac begins, and almost none of them tell you which choice they made.

The number, today

The gap between the tropical zodiac and the sidereal one is the ayanāṁśa. Every sidereal system subtracts one. Here they are for 19 August 2026, computed on Swiss Ephemeris:

Widest gap: 2.33°. The Moon covers that in about four and a half hours.

What it actually costs you

We ran ten thousand births between 1955 and 2010 and compared Lahiri against Raman — the two most commonly used in India:

The rāśi is the least affected thing on that list, which is the opposite of what most people worry about.

Those figures are not an assertion. The script that produced them is committed as scripts/ayanamsa_divergence.py with a fixed seed, so it runs to the same result every time.

A real example

Born 1 June 1990, 15:00 IST:

• Lahiri — Moon at 149.098°, which is Leo 29.1°
• Raman — Moon at 150.544°, which is Virgo 0.5°

Different sign. Different lord. A different Moon-sign horoscope for the rest of that person's life. And note the subtlety: both agree the nakṣatra is Uttara-Phalgunī, because that nakṣatra straddles the Leo–Virgo boundary. Sign and nakṣatra do not have to move together.

The part nobody mentions: your daśā dates move

Vimśottarī daśā is calculated from how far the Moon has travelled through its nakṣatra at birth. Shift the Moon by two degrees and you shift that fraction — so you shift the balance of the daśā running at birth, and every date that follows from it.

Across the same ten thousand births, the balance of the birth daśā moved by a median of 20.8 months between Lahiri and Raman. For the worst-affected tenth, it moved by more than 46 months — nearly four years.

So two astrologers can agree on your entire chart, agree on every yoga in it, and still put the start of your Jupiter period almost four years apart. Not because either made an error.

Which one is right?

We are not going to tell you, because it is not settled. The systems differ on which fixed star anchors the zodiac and on when the two zodiacs coincided — a real astronomical question with a real historical dispute behind it, and no amount of confidence on a website resolves it.

What we will tell you is what we use: Lahiri, the Indian government standard. Every chart this site computes reports it — the field is right there in the API response as ayanamsa: "lahiri", with the numeric value for your birth date beside it.

How to check what an app uses

Ask it. If it cannot tell you, it has made the choice for you and hidden it — and if your Moon is within about two degrees of a sign boundary, that hidden choice decided your Moon sign.

The reason your two apps disagree is almost never that one of them is wrong. It is that one of them did not say.

Sources

• Ayanāṁśa values, 19 Aug 2026 12:00 UT — Swiss Ephemeris — Fagan/Bradley 25°06′44″ · Lahiri 24°13′45″ · True Citra 24°12′36″ · KP 24°07′56″ · Yukteshwar 22°51′03″ · Raman 22°46′58″
• Divergence rates, 10,000 births 1955–2010 — Lahiri vs Raman — rāśi 4.64%, nakṣatra 10.04%, pāda 42.59%; birth-daśā balance median 20.8 months, 90th percentile 46.6 months — reproduce with scripts/ayanamsa_divergence.py (fixed seed)
• The worked example — 1990-06-01 15:00 IST — Lahiri 149.098° (Leo 29.1°), Raman 150.544° (Virgo 0.5°); both Uttara-Phalgunī
• What this site uses, published in every response — GET /api/v2/kundli — meta.ayanamsa = 'lahiri', with meta.ayanamsa_value for the birth date

Check it yourself — free

• Your chart, with the ayanāṁśa stated
• Your Moon sign and nakṣatra
• Your Vimśottarī daśā timeline

All articles · VedicSpace — every number computed on the Swiss-Ephemeris engine (Lahiri), every rule named to its classical source.