There's 1 flaw in this type of analysis and that is an unmeasurable behavioural flaw. Let me put it this way: If this post of "how to be successful on HN" hit

There's 1 flaw in this type of analysis and that is an unmeasurable behavioural flaw. Let me put it this way: If this post of "how to be successful on HN" hits critical mass and gets 400 votes, lingers for an entire 24 hours on the front page and gets loads of clicks (and it contains definitive evidence of "when", "what", "how" to post), it will lead to the opposite happening... How? Well, hundreds of folks will try it out, increasing the "new" rate significantly on the "best days/times to link", making it even harder for anybody to upvote your article, cause the duration on the first page of "new" will be a lot less than normal. The real secret to getting popular on HN is to form voting-rings and upvote your buddies posts ("you scratch my back, I scratch yours"). Alternatively, you can just buddy up with the people who run HN and get them to manipulate your posting by allowing multiple levels of rigging (I've read somewhere that they "allow good posts that were missed to be brought back to life", which makes you wonder what the definition of a "good post" is). Statistical obscurity...