Google's John Mueller confirmed on Twitter that GoogleBot, Google's web crawler, generally does not pass referrer data because it typically only directly accesses URLs. So GoogleBot will not click from page to page on your web site but rather gather the URLs on a web page and visit those URLs at some point directly.
If you visit a page directly, it passes no referrer data.
John did add that "for embedded content (scripts, css, etc) I suspect we include the webpage's referrer." He added you should check your server logs to see if that is true or not.
Here are the tweets:
For webpages, yes. For embedded content (scripts, css, etc) I suspect we include the webpage's referrer (check your logs :-))
— John ☆.o(≧▽≦)o.☆ (@JohnMu) August 10, 2017
Googlebot renders the page as it loads, like it would in a browser. It generally doesn't click around.
— John ☆.o(≧▽≦)o.☆ (@JohnMu) August 9, 2017
Forum discussion at Twitter.
