If you look at my submission history of my blog then I think it's clear that HN likes things that are original and/or well thought out. My weaker blog posts go nowhere, but ones that are detailed make it. So, if there's a formula for appearing on HN, it's write something original and/or deep.
Which can happen very quickly. Sometimes I am not certain whether a posting just isn't interesting, or the wind is so strong the voice gets lost in the roar.
I can think of times when I've been the single up-vote and/or single
comment on a submission of yours. The most recent one that comes to mind
easily is your post on a claim about total memory in ancient computers
(I think it was 53K in 1953 if my memory serves).
It wasn't that the post lacked detail or depth, instead, it's a problem
of varying interests; few people have an interest in ancient computer
system design and history, let alone know what "mercury delay line"
memory is.
When you combine a more esoteric interest with the fast queue of the HN
'/newest' page, the result is an exposure, visibility, and discovery
problem, rather than any real flaw in your content.
Also, robertjmoore, thanks for a great article with data.