Does Google Rank on Traffic Patterns?
An interesting thread over at WMW brings up the question of whether Google uses browsing patterns collected by the toolbar in the ranking algorithm.
There's an interesting range of opinions, from thought provoking to... tin foil hat provoking.
Many names seem to think it is the case -- Tabke for one -- though a few others disagree. If it is fact, the effect on SEO is fairly obvious, though I don't see many people considering that. On the black hat side, it also seems exploitable.
- Y! MyWeb

Noted it a while back
I first mentioned it on 1/13/2005 over at seo chat, and then updated it a year later on a seo scoop post. I called it a kooky theory, but I've since tracked it on other sites, and am at least fairly confident that it holds some merit. The details of it (exactly how) though still elude me.
Yes.
It's not a new thing.
not happening
At the Webmasterworld Pubcon conference in November of last year, Matt Cutts literally said that no data from the Google Toolbar were used to influence the ranking. I've blogged about it: http://lvb.net/item/2026 , see quote 6.
personalized
Of course if you switch on the (experimental) "personalized search results" in Google (you have to be logged on for that), the ranking is influenced by your own clicking-on-search-results behaviour, but you knew that already.
I think it is silly to
I think it is silly to assume they don't track user acceptance somehow. Why WOULDN'T they do that?
Now obviously the algorithms are not traffic pattern only based (or Seth probably wouldn't be complaining about tea results)
I thought econman's post on page 2 sounds good
So maybe a certain amount of trusted user feedback of some sort is needed in addition to link popularity before a site can rank for highly competitive queries.