The model already found titles that score higher. Try one.
I built a deliberately imperfect predictor for HN submissions using ~150k historical posts. Paste a title, URL, and text, and it estimates score/front-page odds while showing similar posts and feature contributions. I mostly use it as a sanity check: does this title look like something HN has historically ignored, or something with a chance?
ForesynWanna keep in touch?
Built this solo over a weekend. Soft-launching before the HN post on Monday. If you scored a draft and the prediction either nailed it or whiffed, I want to know.