How many OBL renders a high PR page useless?

It depends on the PR really. A PR 6 has a lot more pagerank to share than a PR 2.

Generally, anything over 40-50, and I'm not wasting my time. For a PR 1 or 2, I might stretch it out to 25, but really would prefer 10 or less.

Best question I have seen asked on this board for some time. Good job thinking. SO many people either forget or have no idea that OBL affects a page's juice just as much as PR. Obviously none of us have the algo sitting in front of us to know. Like MikeF said it depends on the PR because PR is split or divided by the amount of links . So A PR6 has more to give.

I will just say this . I have in the last few weeks been able to correct some peoples drops in the latest Panda updates with just a PR3 and two PR2 links. However they were extremely low OBL. Less than 10 links. I have also wondered about two other things.

1) Since Google values contextual links as votes that are editorial recommendations (and hpBL seem to confirm this) then why wouldn't Google track the amount of those as well and not just overall links? Too many then its a less impressive recommendation
.
2) People have a way of saying that links that roll off the page still count as having once been on the High Pr page. I really can't see this effect lasting forever as claimed. Since I spend a good amount of time looking at backlinks I can tell you that pages that lose their backlins soon show as missing links and the sites that got those links and lost them often LOSE their PR. Google is cagey with sharing their link info but you can definitely see that with Yahoo and other search engines and the fact that the PR (which is Googles data) from a site drops at the next update after those links go missing tells me this teaching out there is not true.

I think Google does track the ratio of contextual to non-contextual links, they probably just give more weight to the contextual links and less to the non, but this is just my guess.
.
I think you're spot on, if you get a link on a PR6 blog home page and the post rolls off into the archive that link will most definitley lose it's weight after the site is indexed again.

I wasn't talking about comment links mate, I was asking about contextual links where the link is embedded in the article on a high PR page.

Blog comments are not a priority in my backlink plan, therey're just too unreliable. One minute you can have a link on a PR6 page with 10 OBL, the next minute theres 70 OBL and your link just lost it's value.

I do blog commenting, but only for good measure.

I also doubt that comment links will be of any value from now on, it's just too easy to get them and naturally they've been abused, I think Google will soon adjust their algorythm to ignore comment links, wouldn't you?

From my experience, relevence is completley unimportant. This may change in the future, but for now this is the case.

I've managed to easily rank a website in the property niche using only backlinks from completley irrelevent pages.

I can't answer that one because I haven't got a clue mate, we can only guess. Do both!

Also when checking for OBL I'm using SeoQuake and clicking on "Page info" for that particular page and it shows "External links: some number (# no follow)"? Should I go based on those numbers when figuring out how many OBL there are on that page?

I click on "Ext. links" and it shows me the number of dofollow and nofollow OBL.

I see a lot of numbers being thrown around but has anybody done any actual testing on this or are all of you just going by your "feelings"?

I have tested this using links from my network a few different times and I have seen no difference between 10-15 obls and 90-95 obls (I keep them all under 100 so no testing with higher numbers).

This was linking from the same sites with at different times to almost identical sites in the same niche for the same keywords. I tested three pairs of sites with three keywords each and the results I got were within a three places for six out of the nine and the other three were with six places. Sometimes the higher obls resulted in a higher rank and sometimes the lower did pretty much as an even split.

My testing does not seem to backup how some of you are viewing the obl numbers. While I am sure that Google does discount links from high obl pages the numbers are higher than a lot of you are thinking. Now this only pertains to contextual blog posts and may or may not be the same for blog comments as I did not test the comments.


On another note on getting high PR blog comments that are not going to be over run with a huge number comments later on.

Look for high PR sites that have quite a few high PR interior pages where they close the comments after a certain time period (say two weeks). Quite a few of these are unmoderated or in most cases loosely moderated so they are fairly easy to get your comments to stick even when automating as long as you stay way from the viagra type niches and obvious spam comments. While these pages do not have PR now you will find out that a high percentage of them will end up having a decent amount of PR and the obls are going to be pretty low. I have been working on this for quite awhile now and have some very nice high PR comment links on pages with low obls including a PR 7 with just 12 comments. Save the sites you find that meet the correct criteria and then check them for new posts to comment on regularly and you can build yourself a nice list of sites to work from.

Well PR is only loosely related to serps so going from PR being divided by obls and how well they work for ranking a particular search term is not a connection you can really make.

The sites that I was linking out from were mine so I completely controlled the content, placement and how long the links were active.

I placed the articles as the first post on the site, the content in each post was pretty much the same, I use articles that were spun from the same original article, and once the links went live I left the sites alone for 10 weeks without any changes both on the incoming and outgoing.

The links were coming from the exact same sites in both tests so while I agree that all PR measurements are not accurate since I used the same sites the PR should have been pretty close since I stopped building any links to the sites over a month before the first test.

All of the test domains were two months old when I started each test. I purchased the first set two months before the first test and the second set two months before the second test.

All of the outgoing domains were 2+ years old. Now there was a few months between tests but the domains averaged over 4 years old so a few months should really have had a minimal effect on aged domains.

All the links were indexed in Google, I sent the same set 10 bookmarks to each post page (not the main page) and they were all indexed with a couple of days on both the main pages and the post pages. I checked them all using scrapebox using a site: search in Google

While I realize you can not control everything I did try to make all the factors I could as similar as possible. The main factor I could not control of course what others were doing with their sites.


However even though the conditions were not exactly the same if there was such a drop in the worth of links because of the obls then there should have been a marked difference in the results as during the second test the pages had 7-9 times more obls. The results were just to close to mean anything other than the obls are not as important as a lot of people think they are, especially with lower numbers. Now is there a point at which they become a problem? I am sure there is but my testing and subsequent ranking results show me that it is somewhere above 100. How far above I have no idea.