How Hacker News ranking works: scoring, controversy, and penalties (2013)

Hacker News ranking is no longer automatically penalized for having 'NSA' in the title.

A 2013 analysis by Ken Shirriff reveals how Hacker News ranks articles, detailing a scoring system based on upvotes and time, with a significant factor called "gravity" causing scores to decay. The study found that the published algorithm was largely accurate but subject to substantial "tweaking" through penalties. Approximately 20% of front-page stories received penalties. Articles with "NSA" in the title were automatically penalized, as were "controversial" stories that exceeded 40 comments. Certain domains, including popular news sites and tech platforms like GitHub and Medium, also appeared to be automatically penalized. Penalties could drastically reduce an article's visibility, with some dropping off the front page entirely. The ranking system reranks stories periodically, not constantly, and page caching can also affect immediate rankings. This analysis aimed to demystify the "invisible factors" influencing story visibility on the platform.

The analysis highlights that Hacker News actively manipulates rankings beyond simple vote counts, employing penalties to manage content visibility. This suggests a platform that prioritizes certain types of discussion or aims to prevent overwhelming popularity from dominating the front page. The existence of automatic penalties for specific keywords and domains indicates a deliberate editorial strategy to curate the user experience, rather than a purely meritocratic system based solely on user engagement.

This nuanced approach to ranking means that while user votes are crucial, the platform's underlying algorithm can significantly alter an article's reach. For content creators and participants on Hacker News, understanding these hidden factors, like the "controversy" penalty, is essential for optimizing visibility. The system's design implies a trade-off between unfiltered user democracy and platform-guided content moderation to maintain a specific community environment.