[thecoffman]

This seems pretty much in line with my own numbers. I had a post get to the front page for about ~36 hours and held one of the top few spots for a couple hours.

Over approximately 48 hours, I got ~35k uniques, approximately 18k were from Hacker News directly, with the bulk of the remainder coming from Twitter. Many of those twitter referrals came from people RTing HN aggregator feeds and the like though, so in reality probably more than that 18k number can be attributed to Hacker News.

[GrantS]

To provide another data point, I had a top story one day in January of this year, with about 5,000 visitors linking straight from Hacker News and about 10,000 total unique visits for the day.

I wonder if the variance is primarily due to the topic itself, day of the week, or if the community has grown significantly in recent weeks.

[thecoffman]

That's a very good question, it may have to do with time of day or overall business of HN at the time the story peaked. For instance - my story held that position for that long with a little over 100 votes IIRC whereas the submission in this story has well over 200. Naturally one would expect the 200 vote submission to get more traffic despite having a similar frontpage lifespan. It seems as if position isn't the only determining factor in pageviews, time of day is also significant.

All of this is fairly obvious when you think about it, but nice to see data to back it up.