Welcome to WebmasterWorld Guest from 18.104.22.168
Another is --- you'd really be amazed at how many people type in a URL or pull it down from their browser's list of previous sites visited. If you can divide visitors into those who've been to your site before and those who haven't, you might find far more blank referrers for the first group than the second.
There's an item here in this forum that goes into this in more depth.
The referrer field should not be trusted but it should not be ignored either. Like everything in web analytics, there is going to be noise in the data. I would instead say "blank referrer fields should not be trusted." But work with what you have.
[edited by: McElvoy at 2:56 pm (utc) on Jan. 14, 2007]