Ashtakoota Guna Milan: the 36 points, one at a time

Guna Milan is a score out of 36 that compares two birth charts before a marriage. It is built almost entirely from one thing in each chart: where the Moon stood at birth. Eight separate tests, called kootas, each read the two Moons in a different way and each carry a different number of points. Add the eight and you have the number the family is about to argue over.

The eight are worth 1, 2, 3, 4, 5, 6, 7 and 8 points in that order, which sums to 36. The weighting is the whole design. Nadi alone carries 8 points, more than the first three tests put together, so a single failed test can move a match from comfortable to borderline while seven others pass.

Score your own pair on the free calculator. It takes two dates of birth, and it takes times and places if you have them. Everything below explains what it returns and how much weight each part of it deserves.

What each koota reads, and from what

A nakshatra, the 13°20′ arc of sky that the Moon occupies at a given moment, is the finer of the two coordinates. Twenty-seven of them cover the circle, against twelve signs. Four of the eight kootas read the nakshatra rather than the sign. They are Tara, Yoni, Gana and Nadi, and between them they carry 21 of the 36 points, which is why an approximate birth time hurts this score more than it hurts most other calculations.

Two of the eight are asymmetric, meaning they give a different answer if you swap the two people. Varna compares ranks and awards its point only when the man's rank is at or above the woman's. Gana penalises a Manushya man matched to a Rakshasa woman more heavily than the reverse. The calculator therefore asks who is person 1 and who is person 2, and the answer changes if you enter them the other way round.

How high a score actually is

The number 18 is quoted everywhere as the pass mark, and almost nobody quotes what share of pairs clear it. Here it is, computed rather than asserted.

The zodiac divides into 108 padas, each an equal 3°20′ arc, and each pada fixes both a sign and a nakshatra. Running every one of the 11,664 ordered pada pairs through the same eight calculators the live tool uses gives the complete distribution of the 36-point score over all possible Moon positions.

Three quarters of all possible pairings clear 18. The traditional pass mark sits below the median of 23.5 and below the average of 22.26, so clearing it is the common case rather than the good one. If a family is told "we got 19, that clears the minimum", the honest reading is that the pair scored below what more than half of all possible pairings would score.

An independent check agrees. Eight hundred synthetic births between 1985 and 2005, across ten Indian cities in Indian Standard Time, computed on the Lahiri sidereal ephemeris and paired 3,000 ways, produced a mean of 22.24 and a pass rate of 76.1%. The two methods differ in the second decimal place, which is what you want from a combinatorial count and a real ephemeris sample answering the same question.

Nobody scores 36

The ceiling is 35, not 36, and only 16 of the 11,664 pada pairs reach it, which is 0.14%. Every one of those 16 is short by exactly one point, and the point is always Gana's.

The reason is a collision between two of the tests. Yoni awards its full 4 points only when both partners carry the same animal in opposite sexes, which pins the two nakshatras to one of 26 specific combinations. Of those 26, six also give a matching Gana and different Nadis, and all six of them fall at a Tara distance worth 1.5 rather than 3. The four tests that read the nakshatra cannot all pay out together. So a matchmaker quoting 36 of 36 is quoting a number these rules do not produce.

The eight, in order

1. Varna, 1 point

Each Moon sign carries a rank: Brahmin, Kshatriya, Vaishya, Shudra, in descending order. Water signs are Brahmin, fire signs Kshatriya, earth signs Vaishya, air signs Shudra. The point is awarded when the man's rank is equal to or above the woman's, and withheld when it is below.

Across all pairs the point is awarded 62.5% of the time and withheld 37.5% of the time. It is one point out of 36, so losing it moves a total from 23 to 22 and moves nothing else. Families sometimes read a failed Varna as a statement about status, which the rule does not support: it reads two Moon signs and nothing about either person's background.

2. Vashya, 2 points

Each Moon sign belongs to one of five classes: Chatushpada (four-footed), Manava (human), Jalachara (water-dwelling), Vanachara (forest-dwelling) and Keeta (crawling). Same class scores the full 2. Chatushpada with Manava, and Jalachara with Manava, score 1. Chatushpada with Jalachara, and Vanachara with Chatushpada, score 0.5. Everything else scores 0.

Full points come up 27.8% of the time, zero comes up 25.0% of the time, and the mean is 0.97 of 2. Scorpio is the sign that costs the most here, because it is the only Keeta sign and Keeta scores 0 against every class except itself.

3. Tara, 3 points

Tara, the count of nakshatras from one partner's birth star to the other's, is taken in both directions. Count forward from her star to his, reduce by nine, and read the remainder: 1 Janma, 2 Sampat, 3 Vipat, 4 Kshema, 5 Pratyari, 6 Sadhana, 7 Vadha, 8 Mitra, 9 Param Mitra. Six of the nine are favourable; Vipat, Pratyari and Vadha are not. Then count the other way and read that remainder too.

Both directions favourable scores 3. One direction favourable scores 1.5. Neither scores 0.

Zero never happens. Across all 11,664 pairs, 33.3% score 3 and the remaining 66.7% score 1.5, and no pair anywhere scores 0. The two counts are locked together by the arithmetic, so the case where both directions fail cannot arise. Tara therefore contributes a guaranteed 1.5 points to every match ever computed, and its real range is 1.5 points wide rather than 3.

4. Yoni, 4 points

Each of the 27 nakshatras carries an animal and a sex. Same animal in opposite sexes scores the full 4. Same animal in the same sex scores 3. One of seven sworn-enemy pairings scores 0. Every other combination scores 2.

The seven enemy pairs are Horse and Buffalo, Elephant and Lion, Goat and Monkey, Serpent and Mongoose, Dog and Deer, Cat and Rat, Cow and Tiger.

The full distribution across all 11,664 pairs is 3.6% at 4 points, 3.7% at 3 points, 85.6% at 2 points and 7.1% at 0. Yoni therefore behaves as a 2-point floor with a small chance of a bonus and a small chance of a penalty, and a couple told their Yoni is "only 2 out of 4" has the ordinary result rather than a poor one. Full points are the rarest single outcome anywhere in the eight.

Mongoose is the animal to watch, because Uttara Ashadha is the only nakshatra that carries it, while its sworn enemy Serpent covers two, Rohini and Mrigashira. Every other animal appears twice, once male and once female, so every other nakshatra has one partner somewhere in the list worth the full 4. A Uttara Ashadha Moon has none, and its ceiling on this test is the 3 points it scores against another Uttara Ashadha Moon.

5. Graha Maitri, 5 points

Each Moon sign has a ruling planet. Compare how those two planets regard each other, in both directions.

Full points come up 38.9% of the time and zero only 5.6% of the time, with a mean of 3.00 of 5. Graha Maitri is the friendliest of the high-value tests, largely because seven planets rule twelve signs and five of them rule two signs each, so identical rulers are common.

6. Gana, 6 points

Every nakshatra belongs to one of three natures: Deva (mild, cooperative), Manushya (mixed, worldly) or Rakshasa (forceful, intense). Nine nakshatras sit in each class.

Full points come up 33.3% of the time and zero 33.3% of the time, which makes Gana the most volatile of the eight: a third of pairs collect all six, a third collect none. Six points swinging on a single classification is the largest single-test swing in the system after Nadi, and it is the test most likely to move when a birth time is uncertain, for reasons the last section covers.

7. Bhakoot, 7 points

Count the distance from one Moon sign to the other, both ways. Three distance pairs are held to be adverse: 2 and 12, 5 and 9, 6 and 8. Any other distance scores the full 7.

Half of all possible pairs, exactly 50.0%, land on one of those three adverse distances. That alone should temper how the result is read, because a rule that fires on half of everyone is a wide screen rather than a rare warning.

The rule carries an exception, and the calculator applies it. When the two Moon-sign rulers are mutual friends, or are the same planet, the adverse distance is set aside and the full 7 points are awarded anyway. The exception rescues 47.2% of the pairs that trip the distance rule, so Bhakoot ends at zero for 26.4% of pairs rather than 50%. Nearly half of all Bhakoot Doshas quoted from a raw distance count would be cancelled by a rule that sits in the same tradition.

Deeper reading: Bhakoot Dosha in kundli matching.

8. Nadi, 8 points

Each nakshatra belongs to one of three constitutional currents: Aadi (Vata), Madhya (Pitta) and Antya (Kapha). Different currents score the full 8. The same current scores 0, and that is the outcome called Nadi Dosha.

Same-Nadi pairs are 33.3% of all possibilities, so one match in three carries this result. Eight points is 22% of the entire scale, which is why a single Nadi failure drops an otherwise strong match below the traditional line so often.

The score itself applies no cancellation to Nadi. A same-Nadi pair loses all eight points in the arithmetic, every time. The classical exceptions, such as the two partners sharing a nakshatra but sitting in different padas, or sharing a Moon sign while differing in nakshatra, are read by a person looking at both charts rather than folded into the number. So the practical instruction is: treat a 0 on Nadi as a prompt to have both charts read together, and treat the total with 0 on Nadi as understating the pair by up to eight points until that reading is done.

Deeper reading: Nadi Dosha in kundli matching.

Where the points usually go

Two readings fall out of that table. The tests that carry the most points are also the ones most likely to pay in full: Bhakoot pays 73.6% of the time and Nadi 66.7%, which is why the average total sits at 22.26 rather than near the midpoint of 18. And the tests that look most dramatic in conversation, Yoni and Gana, are the two whose typical result is least informative: Yoni gives almost everyone a flat 2, and Gana gives a third of everyone 6 and another third 0.

A worked example, all the way through

Two invented birth records, computed live.

• Person A (entered as person 1): 12 December 1997, 19:25, Pune
• Person B (entered as person 2): 3 February 1994, 17:45, Jaipur

Person A's Moon is in Taurus, in Krittika. Person B's Moon is in Libra, in Vishakha.

The interesting part of this pair is what happens on either side of a single test. Bhakoot would have cost 7 points on distance alone, and the shared Venus rulership gave all 7 back. Nadi then took 8 away and nothing gave them back. Same pair, same two Moons, two adverse findings, and only one of them has a rule in the arithmetic that answers it.

Both charts also carry Mangal Dosha, and both cancel it. Person A has Mars in Capricorn in the 8th house from a Gemini lagna, the sign rising on the eastern horizon at birth. Mars in Capricorn is exalted, meaning it sits in the sign where it is strongest, and Venus occupies the same sign. Person B has Mars in Capricorn in the 7th from a Cancer lagna and the 4th from the Moon, with the same two cancelling conditions. The verdict for each is Manglik with the effect cancelled.

Mangal Dosha in matching, both charts

Mangal Dosha, also called Manglik status or Kuja Dosha, is a separate check run beside the 36 points. It contributes nothing to the score and it is often the thing families react to hardest.

A chart is flagged when Mars occupies house 1, 4, 7, 8 or 12 counted from the lagna, or the same houses counted from the Moon. Either reference triggers it, and severity is taken from whichever reference is worse: Mars in the 7th or 8th is High, in the 1st, 4th or 12th is Medium.

Two further references are computed and reported but never allowed to raise the flag on their own. Mars counted from Venus is a modern addition rather than part of the classical rule. The 2nd house belongs to the South Indian house set and not the North Indian one. Both appear in the result as notes, so a reader can see what was considered and set aside.

How common it is

Across 800 charts with known birth times, 66.0% carry the flag. A separate seeded run over 1,200 charts, kept as a standing check on this rule, puts it at 67.1%, with the lagna alone at 42.1% and the Moon alone at 42.7%. Two references, each covering 5 of 12 houses, produce a union near two thirds.

A finding that fits two thirds of people cannot grade a marriage on its own, and the tradition seems to have known it, which is why the cancellation literature around this one finding is so large.

What cancels it

Four conditions are checked, and they are not equal.

One strong condition cancels by itself. Two weak conditions together also cancel. A single weak condition mitigates without cancelling.

Of the 528 flagged charts, 25.6% are cancelled outright, which takes the effective rate from 66.0% down to about 49%. The 1,200-chart run puts the rate after cancellations at 49.7%, so the two agree.

What it means for a pair

Pairing the 800 charts 3,000 ways gives the numbers that actually matter to a family comparing two kundlis.

Nearly nine pairs in ten have at least one flagged partner, which is why the flag alone tells a family almost nothing. The traditional resolution, matching one Manglik to another so the two conditions offset, is available to the 44.0% where both are flagged. The situation that genuinely needs a reading is the 48.5% where one partner carries an uncancelled flag and the other does not.

Check either chart on the free Mangal Dosha tool. It returns the house count from each reference, the severity, and which cancelling conditions are present by name.

Matching when a birth time is unknown

The most common question asked of this tool is whether it works without a time of birth. It does, with a specific and measurable loss. Here is the size of the loss.

The Moon travels 13.20° in an average day, between 11.78° and 15.36° depending on where it sits in its orbit. A nakshatra spans 13°20′, which is 13.33°. So the Moon crosses roughly one nakshatra boundary per day, and a guess at the hour is a guess at which side of that boundary a birth falls on.

When a time is missing, the calculator uses 12:00 noon as a placeholder and declines to assert a rising sign, because a wrong lagna is worse than none. Measuring 500 births at their true time and again at the noon placeholder, against 500 fixed partners, gives this.

The largest single move observed was 17 points on the same pair.

The worked example above shows the mechanism in one case. Recomputed with both times replaced by noon, Person B's Moon shifts from Vishakha to Swati. Vishakha is Rakshasa; Swati is Deva. Gana falls from 6 to 0, the total falls from 23.5 to 17.5, and the verdict moves from Very Good Match to Average Match, Proceed with caution. Person A's Manglik flag also disappears, because it came from the 8th house counted from the lagna and there is no lagna to count from once the time is gone. Nothing about the pair changed. Only the assumed hour did.

What to do about it

Seven times in ten, the score without a time is the same score you would get with one, so a timeless match is worth running. What it is not worth doing is treating a 19 or a 17 from a timeless run as a decision, because 7.6% of pairs cross that exact line on the hour alone.

Three practical moves, in order of how much they buy:

1. Find the time to the nearest two hours. A birth certificate, a hospital record, or a parent who remembers whether it was before or after a meal will usually narrow it that far. Two hours moves the Moon about 1.1°, which leaves the nakshatra intact unless the birth sits within roughly a degree of a boundary.
2. Run the score at both ends of the range you are unsure about. If the total is the same at 06:00 and at 18:00, the uncertainty does not matter for this pair and you can stop worrying about it. If it differs, you now know exactly which koota is in question and by how much.
3. Read the Mangal check as incomplete rather than clean. Without a time it comes from the Moon only, and 24% of flagged charts are flagged from the lagna alone. A "no Mangal Dosha" result on a timeless chart means the primary reference was never consulted, and the tool says so in the result rather than reporting a clean chart.

More on this: kundli matching without a birth time.

How much the total should decide

The 36 points read two Moons. They do not read the 7th house, its ruler, where Venus sits in either chart, or what period each person is running. A high total describes two charts that harmonise on eight specific tests, and a low total describes two that do not, and both statements are narrower than the conversation they usually start.

The number is most useful in three places. It is a fast screen when a family is comparing several proposals at once. It is a map of where friction is likely, because the koota that scored zero names the area. And it is a reason to look closer, when the total and the impression the two people have of each other disagree.

For a match already decided, the total matters less than which tests failed. A pair at 20 with all the loss concentrated in Nadi is a different conversation from a pair at 20 that lost points across Vashya, Yoni and Gana. Same number, different chart.

Compute yours

• Free Kundli Matching, the full 36 points with a reason line for each koota and a Mangal check on both charts. No account needed.
• Free Mangal Dosha check for one chart, with the house count from each reference and the cancelling conditions by name.
• Free birth chart if you want the Moon's sign and nakshatra before you match anything.
• Relationship deep-dive, which reads both charts together: the strongest bond between the two, where the friction sits, and what each person's current period is doing to the pair. Requires an account, because it holds both charts.

Related

• Yoni Koota, in full
• Mangal Dosha in matching
• Nadi Dosha in matching
• Bhakoot Dosha in matching
• Minimum gunas required for marriage
• Kundli matching without a birth time

How the numbers on this page were produced

Every figure above comes from the same calculators the live tool runs, with no figures taken from anywhere else.

The complete distribution is exhaustive rather than sampled: all 108 padas paired with all 108 padas, 11,664 ordered pairs, each scored on the eight kootas. A pada is an equal 3°20′ arc, so pairing every pada with every pada covers every combination of sign and nakshatra at equal weight.

The Mangal Dosha rates and the pair-level figures come from 800 charts with random dates between 1985 and 2005, random times of day, and birthplaces drawn from ten Indian cities in Indian Standard Time, each computed on the Lahiri sidereal ephemeris. Those 800 were paired 3,000 ways. The run is seeded, so it reproduces.

The birth-time figures come from 500 further births, each scored twice against a fixed partner: once at its true time and once at the 12:00 placeholder used when a time is unknown.

The dates are synthetic. The astrology is the same code that answers a live request.