Yesterday we posted one of our open source projects to HN, and ended up on the front page for most of the day. The biggest complaint in the comments was that we had chosen a bad title for our posting. Yet the post was very successful, it lead to 10k+ views, and 700 stars of the project on GitHub. So would we have done even better with a better title, or was the bad title part of the secret of our success?
An old advertising formula tells that customer's satisfaction is the quotient of what they actually got and what they expected to get. In other words, if you want to increase satisfaction (in this case upvotes), you need to improve the actual content, or lower expectations. This, of course, assumes that your title is good enough that people will click through and it won't get editted by the mods.
We haven't found a way to test this theory yet, but if you do, let us know!