TheMadScientist - 5:11 am on Mar 1, 2011 (gmt 0)
IMO It's too noisy of a metric ... For exactly why you say: a competitor could do that, just like a site could hit you with 3 pop-ups to change your time on the page before a bounce back.
Bounce rate, page views and time on site factors are really something for the site owner to gauge and try to draw conclusions from, but using it externally doesn't work very well for exactly the reasons you are saying ... And these:
For several years, Matt Cutts, along with other Google spokespeople, have been talking about the fact that "bounce" is a noisy signal. Yea, they measure it but apparently they can't use it for any heavy lifting in the algo.
This only makes sense. In many industries what defines a bounce is purely subjective, and often a bounce means success, in the case of a telephone call or sale delivered via a different domain
A common reason for a quick bounce back to the SERP for me is looking for a specific tidbit, finding it, and then wanting to see corroboration from other sites. So I click a new result, spot some reinforcing or conflicting information, click back to the SERP. Rinse and repeat until I have an idea of the consensus.
I may look at SERPS and open a few results in new tabs one after another. I have FF set not to automatically switch to new tab when opened.
I do pretty much the same when reading a web page. If a (usually in-content) link sounds interesting, I may open it in a new tab, but continue to read the current page (so not to interrupt my train of thoughts). Then once finished with the current page, I would go to these newly opened tabs.
These types of behaviors make it so as far as search results go the signals have to much 'noise' (in other words, 'no tellin why a visit went the way it did or what it means' scenarios) to use reliably. (Or so Google's engineers and others keep saying.)
IMO they're something to worry about internally, but not as ranking signals...