Nicely written post! I've had this issue when trying to focus on just the content of the page.
It would be interesting to see if you can find any deeper filtering ideas from other browsers - like Firefox has a "reader view" that strips away cruft from pages. Or even looking at the source code of some open-source feed reader apps like Flym - they do a great job of scraping off content and caching it online in a searchable format.
Thank you for the feedback. I intend to go back and try and find a better basis for the mathematics I used.