Nadi dosha: one in three possible pairings

Nadi dosha is found in 33.3% of all possible pairings of birth star and pada, which is close to one in three. A refusal on nadi usually arrives as a single sentence. The nadis are the same, so the match is not advisable. What that sentence never carries is how often the same table returns that result. Across all 11,664 ordered pairings, a third of them fail this one component, and they fail it before anybody has looked at a birth time, a planet or a person.

How common is nadi dosha, really?

It occurs in one third of equally weighted possible birth-star pairings under this table. The twenty-seven nakshatras are sorted into three nadis of nine stars each. Nadi koota pays its 8 points when the two birth stars sit in different groups and pays nothing when they share one. Three equal groups, drawn independently with equal probability, land together a third of the time.

That is arithmetic before it is a measurement, which is unusual for anything in this field and worth taking advantage of. Three groups, each holding a ninth of the combinations, sums to a third. You can check it on paper in under a minute. The figure above is the same answer reached the long way: we ran every one of the 11,664 ordered pairings through the scorer and counted the zeros. There are exactly 108 nakshatra-pada positions, so squaring that gives the whole population rather than a sample of it, and a second run cannot come out differently.

One third of possible pairings receive this finding. That frequency does not establish what it predicts about a marriage.

Two separate checks follow. First, verify the birth stars and their Nadi assignments: an input error or a birth near a nakshatra boundary can change the lookup. A previous report is not automatically correct merely because the rule is simple. Second, ask how the astrologer used the result. A same-Nadi finding may indeed have been their sole reason for refusal. Other practitioners may consider a cancellation, the rest of the score or a different matching method. The base rate cannot tell us which reasoning a particular practitioner used.

Be careful about what the arithmetic does and does not establish. A common finding is not automatically irrelevant, and a rare finding is not automatically important. To assess predictive value, a study would need observed couples, clearly defined outcomes and an appropriate comparison, not just a count of table outputs. Our enumeration contains none of that outcome evidence. It can challenge a claim that the table result is structurally rare, but it cannot validate or refute a prediction about these two people. Nor can it tell us how many people in India receive the finding: actual birth distributions and partner selection are not measured here.

Two limits on the figure itself, stated plainly. It is computed across the space of possible nakshatra-pada pairings and not over a population of married Indian couples, so it is a structural rate and not a demographic one. And this site does not predict outcomes for anyone, so nothing here says what will or will not happen in a marriage; that is deliberate and permanent.

What is a nadi, and how does a birth star get one?

Nadi is a three-way classification of the nakshatra itself, fixed and published, and the person born under it never enters the calculation. The three groups are Adi, Madhya and Antya — beginning, middle and end. Adi holds Ashwini, Ardra, Punarvasu, Uttara Phalguni, Hasta, Jyeshtha, Mula, Shatabhisha and Purva Bhadrapada. Madhya holds Bharani, Mrigashira, Pushya, Purva Phalguni, Chitra, Anuradha, Purva Ashadha, Dhanishta and Uttara Bhadrapada. Antya holds Krittika, Rohini, Ashlesha, Magha, Swati, Vishakha, Uttara Ashadha, Shravana and Revati.

Notice how the groups are laid out. They do not run in blocks of nine consecutive stars. They cycle out and fold back, six stars at a time: Ashwini to Adi, Bharani to Madhya, Krittika to Antya, then Rohini to Antya again, Mrigashira back to Madhya, Ardra back to Adi, and Punarvasu begins the wave over. The pattern is a fold, not a march. Twenty-seven divided by six is four and a half, so the wave runs four and a half times and still leaves each of the three groups holding exactly nine stars. It is the kind of structure that is obvious in a table and invisible in a sentence, which is one reason people find the assignment arbitrary when they meet it in prose.

The fold defeats the intuition most families arrive with, and that is worth settling before anything else here is applied to a particular pair of charts. Adjacent birth stars can share a nadi — Krittika and Rohini do, and so do Ardra and Punarvasu across the turn — while two stars half a zodiac apart can sit in different groups. How far apart the two stars are tells you nothing at all about whether the koota will pay, so the column has to be looked up and cannot be reasoned out, and a family working it out from the order of the stars will get it wrong about as often as not.

Which star the Moon sat in is itself a birth-time question, though a gentler one than manglik asks. The Moon crosses a nakshatra boundary roughly once a day and a pada boundary every few hours, so a chart drawn from a remembered time and not a recorded one can land on the wrong side of an edge. Planetary positions can be checked against a reference service such as NASA JPL Horizons, but that does not verify a remembered birth time. A report should tell you when a finding sits close to a boundary instead of presenting every verdict with the same confidence.

A widely repeated gloss maps the three nadis onto the Ayurvedic humours: Adi to vata, Madhya to pitta, Antya to kapha. Two people of the same humoural constitution, the reasoning runs, are held to be a poorer physiological pairing than two of different ones. That reading is genuinely old and it is the reason nadi is usually described as the koota concerned with progeny and health.

It is also the reading that carries the least textual support of anything in this post, and it is worth arguing with. The nadi assignment is not derived from anything in the individual — not the ascendant, not a single planet, not the birth time beyond the few hours that fix which star the Moon sat in. Two people born in Ashwini share a nadi whether they are twins or strangers born forty years apart on different continents. A physiological claim resting on a classification that never looks at a body is a claim doing a lot of work on very little.

Eight of the thirty-six points, the heaviest single block

Varna is worth one. Vashya two, tara three, yoni four, graha maitri five, gana six, bhakoot seven. Nadi carries 8, the largest single block in the system. A zero here removes more from a total than any other single failure can, and it does so before the accept line is applied.

Put that against the shape of the distribution and the weighting starts to look decisive rather than merely heavy. The mean Ashtakoota total across the whole population is 20.1 out of thirty-six, and 66.7% of pairings reach the conventional eighteen-point line. Remove eight points from a couple sitting near the average and they land below it. The koota does not merely contribute to the verdict; for a large band of ordinary couples it is the verdict.

Heaviest is still only one of eight. A clean nadi does not make a match, and a failed one does not unmake it; twenty-eight points sit elsewhere.

Which explains something otherwise odd about the tradition. Nadi attracts more documented exceptions than any other koota — eleven separate entries in the rule registry this site runs, against seven for gana. A component heavy enough to decide an outcome on its own is a component that centuries of practitioners have had strong reasons to find grounds for setting aside. The cancellations are not a modern softening. They are as old as the severity, and in several cases older than the point system that made the severity arithmetic.

The verse everybody quotes, and nobody can source

The texts say less than the consultation room implies. The eight-koota scheme with its point weights is set out in the medieval muhurta literature — Muhurta Chintamani is the usual reference, and the 1945 Nirnaya Sagar edition is scanned and readable. The Santhanam translation of the Brihat Parashara Hora Shastra is the other text families are told their refusal comes from.

Now the awkward part. The line practitioners reach for when a family pushes back is rāśyoraikye tu nāḍyaikyaṁ — if the two Moon signs agree, sameness of nadi need not be considered at all. It is half of a couplet whose other clauses do the same job for bhakoot and for gana. Nearly every Indian astrologer can recite it. A survey of the printed editions of Muhurta Chintamani, the Brihat Parashara Hora Shastra and Jataka Parijata failed to find it in any of them. The closest attested parallels are B.V. Raman’s own list of exceptions and the rules carried in the panchanga literature.

That does not make the verse false, and it does not make an astrologer who quotes it dishonest. Plenty of genuine tradition travels orally and reaches print late or never. But a verse everyone quotes and nobody can source is a different kind of object from a citation, and a family being refused a marriage is entitled to know which of the two they have been handed. Our engine cites the text each rule comes from, and where the text cannot be found it says so on the rule rather than quietly promoting it.

There is a broader point buried here about how this material reaches people. The sociological work on contemporary Indian matchmaking — see, for one recent treatment, this study of matchmaking practice in Frontiers in Sociology — describes horoscope compatibility as one filter among several that families apply, alongside caste, education and income. The astrological finding rarely arrives alone, and it is sometimes the most socially acceptable of the reasons available for saying no.

Which conditions do named sources say set nadi dosha aside?

Ten of them, and they split cleanly into those with a named text behind them and those circulating on popular authority alone. These are conditions the sources prescribe. We report which fired and which text each comes from. Whether any of them works is not a claim this site makes or can make.

The ones with a named authority:

• Same Moon sign, different birth star. B.V. Raman, plus the OP, RMP and GBP panchangas. Held by four of the five schools we implement, which makes it the most broadly accepted rule in the set.
• Same birth star, different Moon sign. OP and RMP.
• Same birth star, different pada. RMP and GBP.
• One planet lords both Moon signs. Raman’s Muhurtha, the 1948 Raman Publications edition, exception 5(b), stated explicitly rather than inferred.
• The two Moon-sign lords are friends. Raman, exception 5(c).
• Rasi Kuta and Rajju Kuta both prevail. Raman, exception 5(a). A South Indian formulation, since it is phrased in porutham terms — which is also why it is invisible on a North Indian report that never computes Rajju at all.

And the ones travelling on popular authority:

• Different sub-triads within the same nadi. Jagannath Hora calls this a partial cancellation and declines to say what fraction. Any engine implementing it is choosing a number the source did not give.
• Either party born in one of nine named stars — Rohini, Mrigashira, Ardra, Jyeshtha, Krittika, Pushya, Shravana, Revati or Uttara Bhadrapada. Widely circulated. No classical citation has been located for it.
• A high total, or strong independent marriage yogas. Jagannath Hora again, and this one is circular on its face: nadi is eight of the points being totalled, so a total high enough to trigger the rule is markedly easier to reach when nadi has not failed.

You can see what the split is doing. The sourced rules all test something structural about the two charts — a shared sign, a shared lord, a friendship between lords. The popular rules mostly test the outcome, or hand back a judgement the source never quantified. What a cancellation has to carry before it can be examined at all is set out in the piece on cited cancellations, and the full registry with its sources is browsable in the cancellation reference.

One panchanga refuses every cancellation

The GBP panchanga holds that irrespective of guna score, same nadi should not be accepted. No exception, no exemption, nothing. It is not a heavier penalty or a stricter threshold. It is a position that the ten rules above do not exist as far as this koota is concerned.

That single dissent is the real shape of the disagreement, and it is not the shape most families imagine. Nobody argues about whether two stars share a nadi; the group assignments are identical in every source and the arithmetic is not in dispute. The argument is entirely about what may be done afterwards. One school treats same-nadi as a condition that ten sourced rules can lift. Another treats it as a condition nothing lifts.

Which means a couple can be genuinely, defensibly refused by one astrologer and genuinely, defensibly accepted by another on identical inputs — not because one of them made an error, but because they are working in traditions that answer a prior question differently. Our engine exposes the GBP position as a switch that is off by default, because turning it on overrides ten rules with named sources behind them, and a default that silently does that is a default making an argument on your behalf.

Kerala practice adds another layer by handling much of the same material under the name Rajju, which is why a Kerala report and a Delhi report can appear to be discussing different subjects while examining exactly the same two stars.

What happens when you run the published sample couple?

Priya was born in Ashwini, which is Adi. Arjun was born in Purva Bhadrapada, also Adi. Same group, so the koota scores zero out of 8, and the eight points come off their total before any exception has been looked at.

Now walk the exceptions in order, because this is the part a report should show you and usually does not. Her Moon sits in Mesha, his in Kumbha — different signs, so the same-rashi rule cannot fire. Their birth stars differ, so neither of the two same-star rules applies. Mesha is lorded by Mars and Kumbha by Saturn, so it is not one planet lording both. Are Mars and Saturn friends? Under the standard Parashari table Mars is neutral to Saturn and Saturn is an enemy to Mars, so the friendship rule fails too — and it fails asymmetrically, which is itself worth noticing.

Six sourced rules checked, six not met. Their Ashtakoota total lands at seventeen of thirty-six, one point under the conventional line, and the eight points nadi took are the entire reason. Three of the five traditions refuse the match. Two accept it — and the two that accept are the two that scored them lowest, which is the sort of result that makes sense only once you know that different schools do not share an accept threshold. You can reproduce the whole thing in the nadi tool or see it inside the full score in guna milan.

What should you ask the astrologer who refused?

Three questions, all of which have citable answers, and none of which require you to argue about astrology. Whether you have nadi dosha is not among them. If you were told you do, you do.

• Which cancellation rules do you apply for nadi, and which text do you follow for each? A practitioner working in a tradition can name them. One who cannot is giving you a verdict without a method.
• Do our Moon signs agree, or our birth stars, or do our Moon-sign lords share a friendship? Each of those is a named exception in at least one source, and each is checkable from the two charts you already have.
• Are you applying the GBP position, under which no cancellation is available at all? If so, say it plainly, because it means a second opinion from a different tradition is not disloyalty — it is asking a different question.

If bhakoot came up in the same conversation, and it usually does because it is the other heavy koota, its base rate is even more striking. Ask the same three questions about it.

An astrologer who answers those three has told you which tradition they are working in, and a verdict attached to a named tradition can be set beside another one and compared. A verdict with no method behind it cannot be examined by anybody, including a second astrologer who would like to agree with it. The comparison, not the reassurance, is the thing worth having.