We tried randomly mixing in new stories onto the front page - it was a disaster. The median new submission is pretty crappy, and readers reacted with "how the #@!? did this make the front page", as if we had placed a turd in their breakfast cereal. "We were just trying to test if you would like it" turns out not to be a very popular answer. I've written about this several times in the past, but the only one I can find right now is https://news.ycombinator.com/item?id=21868928.
IIRC, the second-chance pool was our next experiment after that one. It has worked much better. The difference is human judgment or, if you will, taste.
It's not easy to come up with new mechanisms that might help with this problem. Every software mechanism we've tried allows many things through that don't pass muster. Community mechanisms, as soon as you open them up, get overwhelmed by people trying to game them to promote their own stuff. There's a feedback loop with that: the more interesting HN is, the more attractive it is to a high-quality audience, and therefore the more attractive a target it becomes for manipulation, which makes the site less interesting again. So there's a cap on how good it can ever get (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...).
One thing that could be nice is giving established users an occasional nudge to have a look at /new (and a quick explanation of why it's important people do go there). Like, maybe if they haven't been to /new in a week, show a little banner suggesting they give it a go. If you wanted to get really fancy, you could even try to throttle the nudges according to the amount of posts being submitted at the time, so you drive more eyeballs to new when there's more stuff to be reviewed.
I know it took a long time for me to start looking at new, a little nudge would have made me start much earlier.
IIRC, the second-chance pool was our next experiment after that one. It has worked much better. The difference is human judgment or, if you will, taste.
It's not easy to come up with new mechanisms that might help with this problem. Every software mechanism we've tried allows many things through that don't pass muster. Community mechanisms, as soon as you open them up, get overwhelmed by people trying to game them to promote their own stuff. There's a feedback loop with that: the more interesting HN is, the more attractive it is to a high-quality audience, and therefore the more attractive a target it becomes for manipulation, which makes the site less interesting again. So there's a cap on how good it can ever get (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...).