Google PageRank & how to get it.

well, it makes logical sense.

it's a lot easier to coordinate a way to get 10,000 inbounds from two or five sites than it would be to get 10,000 inbounds from 10,000 completely different pages.

 

Read the article Shawn! Any study like this has to be based on some assumptions.

If you want to play with the assumptions the chart in Execl spreadsheet form is downloadable from the article. Plug you own values in and see what happens.

 

These 4,900 backlinks come from 7 domains. Does it matter?

 

Well done!

 

Your theory is wrong!

http://www.webmasterbrain.com/prog/...lr=&ie=UTF-8&as_qdr=all&start=0&sa=N&filter=0

You can see, that gameguru.ru have (believe me - I counted all of them exactly:
- 7 PR7 backlinks
- 186 PR6 backlinks (with 30 links per page in general - it's rating common pages and there are exactly 30 links on it)
- 22 PR5 backlinks
- 312 PR4 backlinks

So do you thinks PR of gameguru.ru should be PR6? Your table says that it should be PR7+?!

 

I think you maybe missing the same thing as Shawn. The chart uses the median value or the range of values for each PR. Nobody knows the actual value of any specific page. Take your PR7 links. The value of these pages could be anything between 152,000 and 823,000. The table calculation is based on the assumption that a PR7 will pass value based on 488,000. If all the PR7s you mention are in the lower portion of their value range you will need a lot more of them.

You also have no idea how close your PR6 ranking is to the top of its range. So rather than your results proving my theory wrong I contend that they could in fact validate my theory.

Why don't you download the chart -- Excel spreadsheet -- and vary some of the numbers and assumptions. For instance, instead of using the median value for a PR, use something much closer to it lower end and see the impact on the number of links required.

I could easily put ups 100 versions of the chart each with a slightly different set of assumptions. The chart can never reflect exactly anyones linking relationships.

 

I couned here http://forums.digitalpoint.com/showthread.php?t=906 all links and their value. If I have the strongest PR6 and all pages have weakest PR4,5,6,7 your theory is wrong - go and check the message!

And I don't understand why you give so many PR from Page to Page if you should:

X multily 85% divide by 50 to get the PR value transfered from Page to Page

 

There is nothing wrong with my mathematics. I do take 85% of the median value and divide it by 50 to get the actual value passed.

But maybe your problem is the sandbox effect. How new are all these links. I certainly haven't allowed for anything but full value given to each backlink.

 

In your opinion, would this algorithm "Add-up" to more relevent searches being done? Do you think this will could provide real world results when applied to search terms?

 

All that I want to say - 5.5 is incorrect ratio between two close pageranks values. I think it's more like 8.

 

I look back at this Infopool article often. I never noticed that the numbers in "Links for PR" are exactly the same as you go down diagonally from Top Left to Bottom Right.

For example...
You would need 16,803 links from PR1 sites to get PR5.
16,803 links from PR2 sites to get PR6.
16,803 links from PR3 sites to get PR7.
16,803 links from PR4 sites to get PR8.

 

I noticed the diagonal thing right off the bat actually. I never bothered to think about it, but I thought it was sort of neat.

anyway... ZanderXML, what you're missing, that article is entirely hypothetical. it's making assumptions, and tons of them. does that make it useless? not at all. is it a chart that you can use to determine what PR you should be at, or how many links you'd need to move up? is it trying to do either of those things? no, no and no.

it may not be on a 5.5 scale. it may not be on 8. it may not be on ANY number; no one said it had to remain consistant. for all we know they could've arbitrarily assigned values for each tier up.

PR5 could be PR4's value ^ 5.5, but PR6 could be PR5's value ^ 210.

something that extreme's not likely, obviously, but nothing says it has to be a completely traditional exponential scale.

 

Nobody says it should be. Article is more then OK.

Another thing I heard that they pass just 65% from the page, not 85% as it was described earlier

 

Not quite. See the way it has been working according to most people who have done tests is in sets of 6 or 7 as bob shows in his article. BT you have to realize that is if there is only ONE outbound link on each of those 6-7 pr7 pages. Any other outbound links decrease the available PR to be passed. The more links the less PR gets passed along and the harder you have to work to get to PR7. It is really not a hard concept to grasp.

If you wish to learn more about how pagerank actually works try this tool: http://www.webworkshop.net/pagerank_calculator.php3

 

The common thought on that is 85%

 

Bob I enjoyed your article. I appreciated your effort in bringing the theoretical towards something practical for Josey and Joe Business person.

I do have 2 questions and a note. I read the table across row "pr1" to the column "links pr for 5" to mean 16,803 pages with "pr1" linking to a target page would give the target page a "pr5." Is this correct? What link structure were you using? A sitemap for the target with a link back to each of the 16,803? Or something else?

My note was that a couple years back I made an experimental site to test allinchor and duplicate content. To my surprise I made a pr 5 page. As I recall the site total was about 2000 pages. I will dig through my notes to provide me info if anyone is interested.

Regards Dan

 

dirtdog1960theone: Yes in the chart it shows that 16803 medium PRx links with 50 outbounds will give you a low value PR(x+4), or as you put it 16803 medium PR1 links with with 50 outbounds will give you a minimum value PR5.

I believe that the article holds that the link structure doesn't matter. No site map needed, no outbound links back to the 16803 pages, this is all inbound links.

 

I wasn't using any specific link structure. For the purposes of this article all links are treated equal and it doesn't matter if they are coming from internal or external pages.

A link is a link is a link. Its PR value is passed to the target page and that is all that matters.

 

Makes me want to get out a little more

 

great article, now I know how to calculate the pagerank already, thanks