Comment by mbreese

I had the same thoughts (high-energy). I would have worded it slightly differently -- more engaging posts.

You could measure in two ways:

1) raw score for the post. Look at the distribution of total scores and remove the low scoring posts. I personally think this will remove more negative posts than positive. (Note: this would be another way to look at this: for the posts with an overall negative sentiment, does the post score more or less).

2) total number of unique people participating in the discussion. The more dynamic posts tend to be more positive, or at least balanced in my mind (might be wrong, but that's my gut feeling).

3) You could also look at the peak rank of the post -- if the post stayed on the front page for more than 1 hour, but this seems more arbitrary and difficult.

I think the idea here is that posts aren't created equal and some have different engagement patterns. What I'd like to know is if a post is skewing negative, does it get more or less traction. What are the incentives for the poster vs the commenter? Both get karma points, but does a commenter get more for being negative vs does a poster get more points for submitting articles expected to have a negative discussion?

There are a number of other questions, like are there keywords that tend to produce negative posts (for example: are posts talking about AWS more positive or negative). Or - are there topics that generally perform better? Are "expected" posts better? Are more "unique" posts better? Are "Show HN" posts more positive than other posts?

I'd be happy to help - info in bio if you're interested.