2. Make assumptions, like that HN's readers are predominantly male. Sure, the probably are, but who knows for sure? If I was asked for my gender when I signed up, I responded randomly. Reliable demographics are simply not available.
3. Test your hypothesis in a way that does not control most variables: Change your name and post on completely different topics. Maybe your male posts were dumber or more interesting than your female posts. Maybe you posted to newer threads with more active commentors. etc. etc. etc. Even your choice of names could bias the study.
4. Have no objective metrics of "niceness". i.e. Evaluate everything with your "gut feeling". (Note: Your gut is biased towards your hypothesis!)
5. Conclude that your assumption was correct based on a statistically insignificant amount of highly suspect data.
I'm not saying your hypothesis couldn't be true. I'm just saying it's going to be a lot more difficult to prove (to even a modest degree of confidence) than you thought it would be. I'm also not saying we should ignore an unproven theory. We act on unproven theories all the time after all.