You'd probably be better off using some sort of time adjusted metric and looking for the median. But even then, what would this information be useful for?
Any sample of users who actually post is going to have higher karma than the universe of HN users, many of whom haven't logged in for years.
Somewhere between the time I had karma 35,000 and 60,000 I had a script that I ran sort-of-daily which captured the top 100 users including the ones who don't have a number posted on that list. I was planning to make a chart of how top user's karma was growing over time but I guess I was never competitive enough to actually make the chart.