I have seen blogs — I can also say forums, but more information exists on the forums which balance out the understanding — have the information incorrect about how link juice is transferred from one page to another, and suggest that linking to another site bleeds link juice off your site. They also ignore relevancy juice by doing so, which is why you to not see this being employed by many sites. The sites, which are seen the most, are doing well in search engines.
… If links leaked juice – digg would be dead …
Simple logic says if linking to other sites caused the site with the links to be reduced in value then a directory would have no value at all. We know that is not the case because often when we search for information we find a directory or a page from an otherwise not relevant site that contains links to relevant pages.
These pages we find are know as authority hubs they contain enough links pointing toward them to make them an authority … in some cases the only authority they have is a single link to them from the site they are on … They are overflowing with relevancy juice because the links they have are about a single topic. They receive this relevancy from their out going links.
There is no registration process to become an authority hub; no add your authority hub url to Google to get a special dispensation given to certain types of content. It is the content itself which makes it an authority hub. Spammers attempt to become authority hubs by scraping search engine listings and linking to those pages in hopes an automated process would work. Google discerns who can get relevancy juice and who can not or what pages they will follow the links from and what pages they will not. Spam is toxic to relevancy.
The leak theory in a nut shell is that each page has a certain amount of link juice and this juice is shared by all links on the page (one must assume equally to understand the theory) by linking to another site or not restricting links to only the site that link juice flows out of the site and is lost to the site.
A similar thought pattern to the idea of leaking link juice would be: Adding additional content – Alone – which links to your home page would result in the home page having more link juice; and result in having a higher rating in the search engines. This theory is easy to test – add 1000 pages to your site all with a site wide link to your home page and check after a google update for changes in page rank.
Digg adds lots of pages but they don’t just add pages – most of these new pages include a vote link to them. For the most part the site relies on links from blogs which bounce link juice off them.
Both theories of linking to another site will reduce the value of the first and adding a page – alone – will increase the value of another site. Ignore the fact that Google is aware of the pages, the site, are one domain. Even if they were not site wide links from one domain, vs one or two links from one domain, to another do not have the effect of an energy juice for the other domain. If we wanted to call something a link juice leak it would be site wide linking.
So this link juice lead theory is a myth! It imagines more power of a site wide link than actually exists. It imagines that links that point to another side dilute the natural site wide link to the home page of the site. … It can harm the ratings of a site because in not linking to relevant material one is ignoring page theme algos. Which in turn result in less visitors and less potential for people who would link to the site naturally to find the site.