Hi All This is just my take on our favourite group guesswork , homework 1, TBPR , which is exported from googles internal PR is solely based on incoming links, 2, Links have weights which are probably dependent on links coming into the page from which the link in question originates from, and the number of link going out of the same page 3, We all observer that the higher PR levels are probably not attained in a linear way, i.e. tis harder to get to TBPR6/7/8/9/10 Based on the above, it must be extremely easy for any very PR page to fall to the next level Afterall, when you have 100,000s links point at 1 page, 1,000s of them per quater will fall out of the google index every quater, directories closing down, webmasters packing it in, websites under new ownership, reorganisations, etc, etc may of the high TBPR pages will also loss some of the link power every quater Ergo, the very high TBPR page must replenish the link power coming into it continously, an infact must stay well ahead of the link drop out rate, its a massive task, except where the TBPR is held by a site that naturally attacts inbound links eg wikipedia, dmoz, goverment sites, universiity sites, statcounter Do directories attract un solicited inbound links ?
My Guess work TB and BL export is done to filter out irrelevant links to update G's database. One don't need to have 1000's of backlinks, instead few high quality relevant links works most. I also feel some Directories provide low value links which reduce the weight of a website.
You are right, and those current links that continue pointing to your domain also might loose PR hence sending less juice your way