Kundli Milan: What the 36 Guna Actually Measure

Kundli Milan — Guna Milan, Ashtakoot, horoscope matching — is the eight-part compatibility test Indian families run before a marriage. Almost every page written about it says the same thing: eight kootas, 36 guna, 18 to pass. That is true and it is not very useful, because it describes the scoring sheet without ever saying what the sheet does when you actually fill it in.

Ashtakoot is unusual among astrological methods in that you can settle those questions by enumeration. Every one of the eight kootas is a function of exactly one thing per person: the sidereal position of the Moon at birth, expressed as a nakshatra and a pada. Twenty-seven nakshatras of four padas each gives 108 distinct positions, so there are 11,664 possible bride-and-groom combinations in total — small enough to compute all of them. Everything below is computed that way, against the same tables our Kundli Matching calculator uses. If you do not know your nakshatra, the Nakshatra Finder will tell you.

Guna Milan uses the Moon and nothing else

This is worth stating plainly because it is the most common misunderstanding about Kundli Milan, and it cuts both ways. It means the method needs very little: not the ascendant, not the Sun sign, not the birth city, not the date. Two people with the same Moon nakshatra and pada score identically against a given partner no matter how different the rest of their charts are. Your birth time matters only insofar as it settles which of the four padas the Moon occupied — a pada is 3°20′ of zodiac, which the Moon crosses in roughly six and a half hours.

It also means Guna Milan is a small slice of what an astrologer looks at in a marriage question. The seventh house, its lord, Venus, Mars, the running dasha — none of them appear in the 36 guna at all. That list is at the end of this article, because it is the part most score sheets leave out.

The eight kootas, and how often each one fails

The eight rows below are the standard table, with one column added: how often each koota scores nothing at all, counted across all 11,664 possible pairs. That column is the interesting one, because it shows the eight tests are nothing like equally hard.

Read the right-hand column and the system stops looking like eight balanced tests. Bhakoot, worth seven guna, fails for exactly half of all possible pairs. Nadi, worth eight, fails for exactly a third. Between them those two account for 15 of the 36 guna and they are the reason most low scores are low. Meanwhile Tara, worth three guna, never scores zero for anybody — and Varna, worth one, fails more often than Nadi does.

36 out of 36 does not happen

The highest total any pair of Moon positions can reach is 35, and only 70 of the 11,664 pairs reach even that. Exactly one birth-star combination gets there without relying on a disputed grade: Mrigashira and Rohini, in 16 of its pada combinations. Every one of those results is short by one guna on Gana.

The reason is structural rather than accidental, and it is easy to state. Yoni awards its four guna only when both birth stars carry the same animal. Gana awards its six only when both carry the same temperament. Nadi awards its eight only when the two constitutions differ. Two of the three heaviest kootas want the charts alike and the third wants them unlike, so the question is whether anything threads all three at once.

Across the 27 birth stars, 3 pairs do:

• Chitra and Vishakha — best total 34.5, short on Tara 1.5 of 3.
• Uttara Phalguni and Uttara Bhadrapada — best total 29, short on Vashya 1 of 2; Tara 1.5 of 3; Graha Maitri 0.5 of 5.
• Hasta and Swati — best total 27.5, short on Tara 1.5 of 3; Bhakoot 0 of 7.

And each of them fails somewhere else, because the fixed spacing that gives two stars a shared animal also fixes the count between them, and that count is what Tara reads. The closest anything comes is Chitra and Vishakha at 34.5 of 36 — a pair that clears Yoni, Gana, Nadi, Bhakoot, Varna, Vashya and Graha Maitri outright and loses half a guna on Tara alone.

So a matching report that hands you 36 out of 36 has not computed Ashtakoot; and a page that describes 33 to 36 as "near-perfect alignment" is describing a band whose top three points are unreachable. If you see a perfect score, that is worth more of your suspicion than a low one.

Tara cannot score zero, and Bhakoot is a coin flip

Two of the eight behave nothing like their headline numbers suggest.

Tara is described everywhere as worth three guna, scoring 3 if neither direction of the count is inauspicious, 1.5 if one is, and 0 if both are. The last case never occurs. The two counts between any pair of stars sum to 29, so their remainders modulo nine sum to 2 — and no two of the three inauspicious remainders, 3, 5 and 7, can sum to 2 modulo nine. Checked over all 11,664 pairs, Tara scores zero for no pair at all. It is not a three-guna koota in practice; it is a koota that awards either 3 or 1.5.

Bhakoot is the opposite. Its rule is all-or-nothing: seven guna unless the two Moon signs stand at one of three forbidden distances from each other — 2/12 Dwirdwadasha, 5/9 Navapanchaka, or 6/8 Shadashtaka. Counted inclusively, the two distances between any pair of signs sum to 14, and that gives seven classes: 1/1 and 7/7, which each cover 12 of the 144 ordered sign pairs because a sign has one self and one opposite, and five classes of 24 in between. The three forbidden distances are all classes of 24, so they take 72 of the 144 and Bhakoot scores zero for exactly half of all possible matches: 5,832 of 11,664, or 50.0%. The single heaviest deduction available in Guna Milan after Nadi is, in effect, a coin toss on where two Moons happen to sit.

The two doshas, and how often the classical exceptions save them

Nadi Dosha and Bhakoot Dosha are the two results that frighten families, and both have recognised cancellations that a score sheet does not apply — it prints the zero and leaves it there.

Both doshas land together for 14.6% of all pairs. That is the combination genuinely worth a consultation, because 15 guna are gone before any other koota is read and the cancellation rules have to be checked against two real charts rather than two nakshatra names.

What the 18-guna threshold is actually doing

Eighteen is a convention. It is not stated as a pass mark in the classical literature, and once you know the distribution it is fairly clear what it was drawn to do: on the conservative reading, 63.6% of all possible matches clear it. A threshold that admits roughly two matches in three is a threshold designed to exclude the genuinely bad ones, not to certify the good ones.

There is a second problem with treating 18 as a line. Six kootas — Varna, Tara, Graha Maitri, Gana, Bhakoot and Nadi, worth 30 of the 36 guna — have rules that are identical across every source. Yoni and Vashya do not: their extremes are agreed, but the intermediate grades vary between lineages. So most matches do not have a single total at all. Only 7.6% of pairs produce one exact number; the rest carry a range up to 3 guna wide.

Which means that for 12.9% of all possible matches — roughly one in eight — the range straddles 18, and whether the match "passes" depends entirely on whose Yoni table the calculator was built from. Taking the most generous reading of every disputed grade, 76.5% of pairs clear 18 rather than 63.6%. Any tool that reports a single number to one decimal place and calls the result a verdict is hiding that.

The four score bands, with the counts behind them

The bands below are the ones in general use. The share of all possible matches falling into each is computed on the conservative reading of the two disputed kootas. The bands were written as though every total were a whole number, and nearly half are not, so a total ending in half a guna is counted in the lower of the two bands it sits between.

Note what the top band looks like once it is counted: 0.8% of pairs, and its stated ceiling of 36 is a point above anything that can occur.

Mangal Dosha is not one of the eight

Manglik status is the objection raised most often in marriage negotiations and it contributes nothing to the 36 guna. It is judged separately, from where Mars sits in each chart — most commonly the first, fourth, seventh, eighth or twelfth house from the ascendant, with the second counted too in several lineages, and with the reference point itself varying between the ascendant, the Moon and Venus depending on whom you ask.

Two things follow. Guna Milan cannot tell you whether either partner is Manglik, so a calculator that reports it alongside eight koota scores is either reading a real chart or inventing the answer. And the exceptions are broad enough — both partners Manglik, Mars in its own sign or exalted, an aspect from Jupiter or Saturn, Mars in a movable sign — that a Manglik verdict without the chart it came from is not information you can act on.

What Ashtakoot does not look at

This is most of a chart, and it is the honest answer to why a 30-guna match can go badly and an 18-guna match can go well:

• The seventh house of either chart, its lord, and any planet sitting in it — the house of marriage itself contributes nothing to the 36 guna.
• Venus and Mars, which between them carry most of what an astrologer reads about attraction, desire and marital friction.
• Mangal Dosha. It is checked separately and is not one of the eight kootas, despite being the objection families raise most often.
• The running Vimshottari dasha of either partner, which decides when a marriage is favoured rather than whether the pair matches.
• Every planet's strength, aspect and dignity — Ashtakoot uses only the Moon, so two people with identical Moons score identically no matter what the rest of their charts say.

How to read your own result

1. Read the eight rows before the total. A score of 22 with full marks on Nadi and Bhakoot is a stronger match than 26 built on a zero in either. The total is a sum, not a summary.
2. Check whether a zero is cancellable. A Bhakoot zero is answerable from the two Moon-sign lords alone, and nearly half of them are. A Nadi zero has three recognised exceptions. Neither will be applied for you by a score sheet.
3. Treat a hostile Yoni pair as the real signal. Only 7.1% of pairs hit one of the seven hostile animal pairings, and unlike Bhakoot it is not a distance artefact. It is the rarest of the large deductions and the one least likely to be noise.
4. Ignore Varna on its own. It fails for 37.5% of pairs and is worth a single guna. No tradition treats it as disqualifying by itself.
5. Ask what the number was computed from. If a tool cannot tell you both nakshatras and both padas, it has not computed Ashtakoot, because those four values are the entire input.

Frequently asked