Why two astrologers gave you opposite answers

Two astrologers can reach opposite verdicts on the same two charts without either one making a mistake. One family now believes the match is unworkable and the other believes it was never a problem, and the natural conclusion is that somebody incompetent is involved. That conclusion is usually wrong, and holding on to it is what keeps a family stuck. What has almost always happened instead is that two documented traditions were applied to the same two birth details, and the traditions disagree with each other far more often than anybody outside the subject expects.

Does a contradiction mean one of them got it wrong?

Not on its own — the two verdicts can both be correct applications of rules that were never reconciled with each other. Kundli matching is not one method. It is a family of methods with shared ancestry, several centuries of independent development, and no arbiter sitting above them.

Consider what an astrologer actually has to decide before the chart is even drawn. Which system: the eight-koota Ashtakoota total, or the ten poruthams used across Tamil Nadu and Kerala. Which value for the ayanamsa, the offset that fixes where the sidereal zodiac starts. Which of the two circulating yoni tables. Which house list for Mars, and which points to judge Mars from. Which cancellation rules to honour, out of a registry that runs to dozens. Every one of those has a defensible answer and a source behind it, and every one of them can move a verdict.

None of those choices is visible on the report. That is the actual problem. A family is handed a conclusion — it does not match — with the entire chain of choices that produced it stripped out. The conclusion looks like a measurement. It is a measurement plus about six decisions.

There is a temptation to resolve this by declaring one tradition the real one and the rest degraded copies. It is a tidy story and it does not survive contact with the texts. The Brihat Parashara Hora Shastra and Jataka Parijata both describe koota material, and neither hands down the modern 36-point arithmetic in the form calculators now use. The scoring convention is later than the rules it scores.

How was the disagreement rate actually measured?

By running every possible pairing rather than sampling some of them. A Moon can occupy exactly 108 nakshatra-pada positions — 27 birth stars, four quarters each. Two people therefore have 11,664 ordered combinations between them, which is small enough to compute end to end in a couple of seconds.

The headline combines four checks: a change across the 18-point threshold between raw and cancellation-adjusted scores; disagreement between school scores at that threshold; disagreement between the selected Ashtakoota and Dashakoota acceptance tests; and a veto raised by a tested configuration. Each pairing is counted once if any check fires. That union contains 56.4% of the enumeration. A veto can enter the union without two schools giving opposite final verdicts, so this is not the rate of five-school verdict disagreement alone.

More than half of the enumerated pairings trigger a tested difference or veto. The individual check explains what was counted.

That number has no error bar, because it is not an estimate. There is no seed, no sample, nothing that would come out differently on a second run. It is a count over the whole population of birth-star combinations, and you can re-derive parts of it by hand. Nadi sorts the 27 stars into three groups of nine, so two stars land in the same group about a third of the time, and the census returns 33.3% for nadi dosha. Bhakoot fails on six of the twelve possible distances between two Moon signs, so it should come out at half, and it does: 50%. Two closed-form checks that the machinery has to reproduce or be wrong.

The manglik question sits outside that figure entirely, because Mars needs actual planetary longitudes rather than birth stars alone. It was measured separately over 2,000 computed charts, and it is the worse number: 61.4% are manglik under one school and completely clean under another. You can put your own details through the manglik checker and see all five verdicts side by side rather than one.

Why do the schools agree on how many couples pass but not on which ones?

Because they differ in tables rather than in severity, and a table change reshuffles the population without shrinking it. Across all pairings, 66.7% clear the 18-point line under a typical reading, and the mean total is 20.1 out of 36. Every one of the five schools lands within a few points of that acceptance share. Read the headline figures alone and you would conclude the schools broadly agree.

They do, about the population. They do not about individuals. Swapping one koota table for another moves some couples up and roughly as many down, so the total barely shifts while the membership of the accepted set changes considerably.

Then there is the boundary. A pass mark is a step function: 18 accepts, 17.5 refuses, and the distance between them is nothing at all. The distribution of totals is dense around the middle, which means a large number of couples sit within a point or two of the line. For those couples a single differing table — one yoni grouping, one convention for scoring friendship between Moon lords — is the whole verdict.

A score near 18 can cross the conventional line after a small table change. A total far from that line will not cross it after the same small change, but that does not make the whole assessment robust: cancellations, separate vetoes and different systems must still be checked. This enumeration does not measure where most real-world refusals occur or how any score predicts compatibility.

Which fork does the most damage?

Not the one everybody argues about. The assumption practically every family arrives with is that the disagreement is about strictness: one astrologer is orthodox and hard to please, another is lenient and finds a way through. That axis is real. The census found it smaller than two others.

The largest is which system is being used at all. A couple assessed on Ashtakoota and the same couple assessed on the ten poruthams are being examined for different things, and the two systems disagreeing with each other is one of the biggest single sources of contradiction in the whole exercise. That is not a subtle methodological point when families now routinely marry across the distance between Ludhiana and Coimbatore. The difference between the two systems deserves its own reading, because it is not a matter of accent.

The second is the reference point for Mars, which almost nobody mentions.

Between the narrowest and the widest documented profile, the share of charts flagged manglik runs from 11.7% under Astro-Vision (serious Kuja dosha only) to 73% under OP / SKU panchangas. Same charts, same planets, different rule. The reference-point problem is worked through properly elsewhere; the short version is that judging Mars from the ascendant alone, or also from the Moon and from Venus, is a decision made before your chart was opened and is rarely written down.

Where does the zodiac start, and does it matter?

It matters at the boundaries, which is where a surprising number of charts sit. Sidereal astrology needs an ayanamsa: a number describing how far the sidereal zodiac has drifted from the tropical one. The values in common use differ by more than a degree. Lahiri is the default almost everywhere; Raman specified his own; a few schools use others again.

A nakshatra is 13°20′ wide and a pada is 3°20′. A degree and a half of disagreement is a bit under half a pada. For most charts that changes nothing — the Moon is comfortably inside its quarter and stays there. For a chart sitting near a boundary it changes the birth star, and if the birth star changes then every koota is recomputed from a different starting point and the report is simply a different report.

The same logic applies to the recorded birth time, and it is a bigger practical problem in India than the ayanamsa is, because a great many birth times are remembered rather than documented. The Moon moves through a pada in roughly two hours; the ascendant changes sign about as often. So “around four in the afternoon” is not a time, it is a window, and a window can contain a boundary. The birth-time sensitivity tool exists to answer the only question worth asking here: does the verdict hold across the window, or does it flip inside it.

None of this is about the accuracy of the underlying astronomy, which is a solved problem. Planetary positions come from JPL’s DE440 ephemeris, and the historical civil-time offsets needed to turn a local clock reading into a universal time come from the IANA time zone database. The planets are not in dispute. What is in dispute is which frame to read them in and which rules to apply once you have.

Why were these traditions never reconciled?

Because for most of their history they were not in contact with each other in the way that would have forced the question. This is the digression, and it earns its place, because it explains why the problem is getting worse rather than better.

A matching tradition only has to be consistent with itself when the families it serves marry within a few districts. A Kerala astrologer applying poruthams and a Punjabi astrologer applying Ashtakoota never had to agree, because they never assessed the same couple. There was no occasion to notice that the two systems count different things, and no institution with any standing to reconcile them if anyone had.

What changed is the marriage market, not the astrology. Search widened through migration, then through matrimonial advertisements, then through matrimonial sites. Banerjee, Duflo, Ghatak and Lafortune’s study of newspaper matrimonial advertisements documents how deliberately those preferences are stated and how the search operates as a market. The NFHS-5 national report records the accompanying shift in when and how people marry. Two systems that never met now meet constantly, inside individual families, with nothing available to adjudicate between them.

So the contradiction a couple runs into is not a sign that the tradition is breaking down. It is a sign that it is being asked a question it was not built to answer: what happens when the two households follow different systems.

If they all contradict each other, is any of it worth taking seriously?

The contradiction is an argument against treating one verdict as final, not an argument for dismissing all of them. This is the objection a sceptical reader raises immediately, and a relative will raise it too, from the opposite direction: if you can find a tradition that says yes, you have proved nothing except that you went looking.

That objection is fair and has a real answer. Shopping for a favourable verdict is asking one question — is there an astrologer who will say yes. Asking what all five documented traditions say about the same two charts is a different question, and it has a different kind of answer. If four of the five refuse, you have learnt that the refusal is broad. If four accept and one refuses, you have learnt that the refusal is narrow and you can name the tradition it comes from. Those are opposite situations and from the outside they look identical, because both arrive as the sentence “it does not match”.

The other half of the answer is that these systems are not arbitrary. Each is internally consistent, most of their rules trace to a printed source, and people have applied them carefully for a long time. When a house list omits the 2nd house it is because the verse behind it omits the 2nd house, not because somebody was careless. A family that follows one of these traditions is not being unreasonable, and telling them they are is the fastest way to end the conversation.

What we hold to is narrower and, we think, harder to argue with: a verdict without its method attached cannot be examined, and anything that cannot be examined should not be the last word on a marriage. Every rule the engine fires carries the text it comes from, which is what makes a reading something you can check rather than something you have to accept.

What does the census not tell you?

It measures whether the verdict changes between traditions, and nothing beyond that. Stating the limits plainly is part of the claim being worth anything.

The star-only figures weight possible nakshatra-pada pairs equally, not according to an observed birth cohort or partner-selection pattern. We have not measured how closely those rates match real couples. The manglik figures come from generated charts spread over dates from 1970 to 2025 and a handful of Indian cities; a real population has a different age and geography profile.

Ayanamsa is not modelled in the headline at all. All five schools were run on Lahiri for the census, even though the Raman reading properly specifies its own value. The 56.4% figure therefore excludes ayanamsa sensitivity. Adding a new check to the same union could not reduce that union, but rerunning the original checks with changed inputs could change their outcomes in either direction. Neither exercise has been measured by this headline.

And the biggest limit is the one a number cannot fix. A counted flag means at least one of the stated checks fired, not necessarily that two final verdicts conflict. It does not mean the difference is material to your family, or that a household should change a decision because a fifth tradition disagreed with the one they follow. That is a family question, and no census answers it. What the census can do is stop a narrow refusal being mistaken for a universal one.

What should you ask the person who gave you the verdict?

Four questions, in this order, and a careful astrologer will be pleased you asked. Which tradition are you applying. Which houses are you counting for Mangal dosha, and from which reference points. Which cancellation rules did you check, and what did each one require. What birth time did you use, and which ayanamsa.

Those four are not a challenge and should not be delivered as one. They are the questions that turn a conclusion back into reasoning. An astrologer working from a lineage answers all four without hesitation, because the answers are simply what they do. If the report came out of software, ask which program — that program made the choices, and its conventions are the tradition your family received whatever lineage they believe they follow.

The specific phrasing matters more than it sounds. “Are you sure?” invites a defensive yes. “Which houses did you count for Mars?” invites a fact. One of those is a conversation and the other is a stand-off, and the family sitting in the room can tell the difference immediately.

When the answers arrive you no longer hold two contradictory verdicts. You hold two clearly labelled readings — this one refused under a southern reading that judges Mars from Venus, that one accepted under a northern reading that does not — and the argument at home changes shape. It stops being about which astrologer erred and becomes about which tradition your household follows. That question has no objectively right answer either, but it is one a family can actually discuss. Running the same details through the full matching tool under all five, or reading what each of the five actually says, is a way to arrive at that conversation with something specific in hand rather than a grievance.