SkyLoka · akandam.com · The evidence

We tested whether the stars predict anything.

Then we published the answer — including the part that doesn't flatter astrology. This page is generated from the same findings.json the app ships. Nothing is edited by hand.

The question

If a planet's sign at birth shaped a life, then across hundreds of accomplished people some signs should appear far more often than chance for the planet that "rules" their field. That is a claim you can actually check. So we checked it.

How the test works

  1. Take the verified register — real, sourced birth dates of documented people. Positions depend on the date, not the clock, so this is honest even where the birth time was never recorded.
  2. Compute each person's sidereal placement (Lahiri) for Sun, Mercury, Venus, Mars, Jupiter, Saturn and Rahu. The Moon is excluded — it moves ~13° a day, so without a birth time its sign is genuinely unknown for most of the register.
  3. Build an era-matched baseline: for each person, 30 "surrogates" jittered ±15 years. This scrambles the planet's phase while keeping the same historical era — so any real signal has to beat the era itself, not just a flat random guess.
  4. Compare observed sign counts to that baseline with a chi-square test per planet, then apply a Benjamini–Hochberg false-discovery-rate correction across all planets tested — because if you test seven planets, one will look "significant" by luck alone.

The result

Loading the published findings…
Planetnχ²pFDRstrongest signsurvives?
…

What it means — plainly

The eye-catching raw numbers — "Saturn in Virgo appears 37% more than expected!" — are exactly the mirage the correction removes. Against an era-matched baseline, across a proper multiple-testing correction, no planetary sign placement holds up. The lifts collapse. That is not a bug in the test; it is the test working.

This is the honest result, and it is why SkyLoka refuses to tell you the stars will make you rich, sick, married or sued. Our own evidence says the placement doesn't predict the outcome. What we will do instead is show you the real sky, name the tradition without pretending it's physics, and let you find the documented people who genuinely share your placements.

An app that had run this test and found nothing would have every incentive to bury it. Publishing it is the whole point.

Reproduce it. The register and every computed placement are public: luminaries.json · placements.json · findings.json. Method & ephemeris: /proof/method. Back to the register: /luminaries · the proof cards: /proof.