Hacker News Reader

Comment by credit_guy | Hacker News Reader

> Obviously astrology does no better than random guessing

Not necessarily. It is entirely plausible that babies born during the Summer months share something in common, and babies born during Winter months share something else. Since astrological signs are 100% correlated with the seasons, one could "guess" with a higher probability than random chance these traits. The causality would obviously not be from stars to babies, but from seasons to babies, but an observational study could end up "proving" that astrology works by identifying these "predictions".

View on HN

This study was a followup to one that just used signs [1], which yielded performance no better than chance. If the summer/winter correlation held, then that study would have shown positive results, which it did not.

There's nothing to say that these star charts or seasons of birth cannot correlate with personality traits or life outcomes. But Astrology, as a practice, is not useful for unwinding those correlations. If you have some alternate system, feel free to propose it -- pump the star charts and personality charts from this study into some AI and see if it can come up with some kind of super-astrology, and then try to replicate it.

I doubt you'd find anything stronger than predicting the answer to the question "were you born in the summer or the winter".

[1] https://www.clearerthinking.org/post/we-tested-the-predictiv...

Also, summer in one half of the world is winter in the other, so the signs do not correlate 100% with seasons, and you'd have to study across geographies.

To give a concrete example: https://en.wikipedia.org/wiki/Relative_age_effect

But whether astrology actually settled on these features would depend on the feedback systems in place, and the granularity of those systems/how predictions were made in reference to a specific person.

> It is entirely plausible that babies born during the Summer months share something in common, and babies born during Winter months share something else.

This would not account for reversal of the seasons in opposite hemispheres...

... or do the northern and southern hemispheres practice a different "western" astrology (!).

90% of humans live in the northern hemisphere, it would still be strongly predictive if true.

Or ever more local differences like when school age cut off is. I would expect some small differences be found due to these impacts, but they would not be universally applicable without generalization.

If schools roll over to the next grade at the start of October, babies born in October will perform better due to being the oldest in their class, but this would not be universal. What I would expect to be universal, is given a school system that does grade division in this way, students right after the division would do better than the ones born right before, as the first group is the oldest in their class and the second group is the youngest.

The extent this pairs with astrology would probably depend purely on population densities between different cultures. If things were fully uniform it would average out, but given they aren't, some residual bias existing wouldn't surprise me.

I predict: people will be warning warm jackets on your birthday

Reply on news.ycombinator.com