Compatibility scores: what one number is built from and what it can answer

A compatibility score is a number, usually out of 100, and the first useful thing to establish is what job it is doing. It is not answering whether a relationship works. It is compressing several dozen chart measurements into one figure so you can sort, compare, and decide where to look first. Compression loses information by definition. What comes out the other end is a summary of the geometry, not a verdict on two people. Moon Shine AI shows a score alongside the reasoning that produced it, and the reasoning is the part worth reading. Once you know which layers a number was built from, you can tell what it is competent to say.

What the number is competent to do

Start with the use, because it settles most of the arguments. A compatibility score is useful in two situations and does very little outside them.

Ordering your attention is the first. A synastry comparison between two charts produces somewhere around forty to sixty contacts once you count aspects between planets, contacts to the angles, and house overlays. Reading all of them in whatever order the software prints them is how people end up with a flat list of statements and no sense of what actually matters in that pair. A weighted score is built by deciding what matters, and a decent implementation shows you that ranking. The number itself is close to a by-product of the sorting.

The other use is comparison. A score on its own has no calibration, since nobody knows whether 68 is a good figure until they have seen a few dozen produced by the same method. Run one method across several relationships you already understand, including ones that ended and ones that lasted, and the numbers start meaning something in that method's own terms. That exercise is also when you find out how much you trust it.

What a score cannot do is answer the question people usually bring to it. Whether to stay, whether to begin, whether this one is worth the trouble. Compression that severe will not carry that kind of weight.

The layers underneath a single figure

Almost every scoring method built by a competent astrologer is layered, meaning different classes of contact are weighted differently before anything gets added up.

Contacts between the lights carry the most. Sun to Moon, Moon to Moon and Sun to Sun, in either direction, often account for something in the region of a third of the total on their own. These are the contacts that show up in how two people live together across years rather than in how they feel during the first month.

Personal planet aspects come next, with Venus and Mars contacts weighted for attraction and Mercury contacts for whether ordinary conversation works. Contacts to the angles, particularly the Ascendant and Descendant, carry real weight when both birth times are reliable, and they are usually discounted or dropped entirely when the times are rounded.

Then there is the balance question. Most methods track the proportion of supportive to challenging contacts, but they differ on what to do with the result. Some subtract for squares and oppositions. Others treat a moderate share of hard contact as a positive, on the reasoning that a pairing with nothing to push against tends not to hold two people's interest for long. Neither treatment is wrong. They are answering different questions about the same geometry.

Why one couple scores 71 in one place and 54 in another

People discover this and conclude that scoring is arbitrary. It is not arbitrary. It is under-specified, which is a different problem with a different remedy.

The variables are easy enough to list once you know where to look. Orb choice decides which contacts count at all, and a method allowing 8 degrees for a Venus to Mars square will find things that a method allowing 4 degrees never sees. Whether house overlays are counted, and whether they are counted in both directions, moves the figure substantially. Some methods fold the composite chart into the total while others keep it out. Weightings differ by design rather than by accident.

The remedy is transparency rather than agreement. A score you can interrogate, where the contributing contacts and their weights are visible, stays usable even when you disagree with the weightings. A bare number from a method that hides its working leaves you nothing to check.

A low number is a map of friction, not a sentence

This one needs stating plainly, because the damage a badly framed score does is real. A low compatibility figure does not mean a relationship is doomed, and it never has.

What a low score reliably indicates is that the geometry between two charts holds more friction than ease. Friction is information. It tells you which areas will require effort, and effort applied knowingly works out very differently from effort applied in confusion about where the difficulty is coming from.

The record is unkind to the doom reading anyway. Long marriages turn up with unremarkable scores constantly, and pairs scoring in the high eighties separate inside two years. A score describes potential sitting in the geometry. Whether that potential gets used depends on factors the chart never measured.

If you are holding a low figure right now for a relationship you are actually in, the useful move is to ignore the number and read what produced it. A total dragged down by Saturn contacts is describing something quite different from one dragged down by Mars contacts, and those two situations call for different kinds of attention.

The problem at the other end

High figures deserve the same scepticism and almost never get it.

A high score means the contacts are predominantly easy, and ease carries its own failure mode. Relationships with very little friction sometimes lack the traction to develop at all, drifting pleasantly for years while nobody decides anything. Two people can also be well matched by geometry and want incompatible lives, which no synastry method measures. Where somebody wants to live, whether they want children, what they are willing to give up. None of that is aspect data, and a high number offers no protection against it.

The practical consequence is that a strong score should change how you read rather than how confident you feel. Ease in the geometry means the difficult material in that relationship is located somewhere the synastry did not look, which is worth knowing in advance.

What no score ever had access to

Several factors sit outside the reach of any compatibility method, and they are usually the ones deciding the outcome.

Intention is one. Two people who have decided to make something work behave differently from two people testing whether it will, and no aspect in either chart distinguishes those cases. Maturity is another, and unlike the natal geometry it changes. The same pair of charts at 22 and at 40 produce an identical score and a completely different relationship.

Context matters just as much and is even less visible. Distance, money, health, family obligation, and how much room each person has in their life at the moment all shape a relationship more directly than a Venus square, and none of it appears in a synastry total. Astrology is one input into the decision. The decision belongs to the people making it.

The score in Moon Shine AI runs out of 100 and is layered the way this page describes, with the lights weighted heaviest, and it is shown next to its reasoning on purpose. Now try the method on charts you can vouch for. Take three relationships whose actual history you know and run each through the same scoring method without looking at the result first. Write down what you expect, then compare. Where the method disagrees with your experience, go and find the contacts responsible for the gap. That tells you more about a set of weightings than any explanation of them, and it shows you how much confidence the number has earned.