Link value

Hello,

I would like to check my assessment with the seo gurus.

Imagine you can get one back link from (i) a pr2 page with 50 outbound links, or one back link from (ii) a pr3 page with 250 outbound links.

I believe you get more pr juice in case (ii); right??

The reason is the logarithm scale (I assume a log10 basis). Because log(100/50)=2*log(1000/250).

So (ii) gets about twice as much juice as (i). Right?

Umm, that sounds right - let's see inserted arbitrary # assignment for outdated and non-value-reflective number, invent mathematical equation, multiply by 2 for no apparent reason. Yep, checks out. Sounds about right to me.

lol
I wish I invented the equation, but Page did this b4 me; in case you forgot, it is here:
http://en.wikipedia.org/wiki/Pagerank

Now it is exponential scale, right.
So that's what I did.
Obviously, everything else being equal

If you want to make a rebuttal, give me a counter equation.

Thank you very much

edit: obviously noone knows the formula for pr; but hopefully there are people in this forum who can use their brain a little.

Such theoretical musings are all well and good when it comes to whiling away the long winter nights, but they're not going to have any real world practical benefits.

As a hypothetical question it necessarily ignores too many important considerations to have any worth in terms of SERPs impact -- even PR impact for that matter.

ok JagNet,

I now realize that I have entered the slippery slope.

1. If I ask the question, there is a reason; the reason is I can choose between (i) and (ii)

2. Ok so I guess noone knows for sure as this is not really testable, right?

3. What is wrong with my formula, then?

4. Or let me reask the same question in a different form; how do u choose a back link between a given pr page, or a higher pr page with much more outbound links.

Yeah, I know. You've got to have something to start with (i just had to do something to amuse myself today).

My point is, basically, PR2 and PR3 are not accurate numbers - it's toolbar PR which is very different from what's used in any algorithmic calculation.

The real question, beyond any formula, is how do you chose (if you have to) which page to get a link from. I think you could develop a formula for this, but have to be much more complex taking into account relevancy between pages where on the list of links yours will appear, the theoreticall possibility of reciprocity (in other words is it possible that you're linking to a site that links to one of those pages). I'd say if you were guaranteed to be the first visible link (to a bot) on either page, both carrying the same relevancy, both as contextual links (rather than in some module), then yes you pick the one with the higher pr.

My experience with single external links like this though, is that position on the page, placement, relevancy, context, etc. can all be more important factors (remember your formula comes from a very dated thesis, there's a lot more going on now).

I still like my first response better, but hopefully this one allows for a little bit better discussion (because I do think it's a good debate to have).

Flip a coin, or get both!

Personally I would be happier with less links but the way it sounds is as if you may be on a "link page" which will be of little value long term.

I'm not dismissing your reason for asking it, and I assumed that you probably were facing the choice between gaining a link from either such source.

There's too many variables to make adequate tests, and you can't really put in place the necessary "controls" that would make tests meaningful.

The principal problem is not with the formula itself, but in the assumption that the formula is the only factor at play. Even if you're trying to calculate purely PR benefits and ignoring the even more complex SERPs benefit, you're still assuming that all links will definitely receive an equal share. I'd be loathed to leap to such assumptions.

I can't envisage myself being in a situation of having to choose between links, but hypothesising for a moment, I'd want to look at the likely factors that would influence my SERPs rather than focussing on PR. Which site is more related to mine? Which list of links is most closely related to mine? Which list of links contains the best sites in the niche? Does one page have a list of links only, whilst the other also contains brief descriptions? Does either site require a link back in return?

Thank you guys,

I am glad to see so much action on my thread. Now, I understand all of the above. I need to give more details, and maybe I will gain some insight. So here are the details:

1/ I have said and repeat it: everything being equal; this is a theoretical question; so do not worry about other factors; in any factor equation, let them all stay fixed, and ONLY change one; this is what I am doing here.

2/ this is a site wide link exchange I am talking about; so I am not sure how relevant the question of link placement is, specially as with css layout, you can play with this a bit; are you saying that you want the link to be read sooner than later by the bot, no matter what?

3/ yes this dilemma is a dilemma I am facing because I only want to site wide link exchange with a select group of sites, hence I can choose, and want to choose at best.

4/ of course these url are both related to my business; this is a repeat of 1/; I am not going to exchange with unrelated url anyway, but this is absolutely not the question of this thread

5/ did I forget something; probably yes, but you got the idea.

Let me repeat for those for practice fast reading:

EVERYTHING BEING EQUAL, do you prefer getting a site wide link exchange with a main page which has pr with 50 outbound links, or with another page which has pr+1 with 5 times more outbounds, i.e. 250.

If you read all this, tyvm.

Sitewide link exchange? Everything being equal, Google will likely see this as a blatant attempt to manipulate SERPs and PR and discount the link anyway.

Since it's sitewide link, then who's to say (ignoring the above for the moment) that it's actually the homepage of each that will be the strongest link? Perhaps there's a PR2 inner page on the PR3 site with zero outbound links?

The trouble with "everything being equal" is that it never is, so any assumptions based on that are always going to be fundamentally flawed. As I said before, that's fine if you're just exercising the mind for a bit, but not so good if you're basing your back link strategy on it.

ok,

now this is something else which I am also trying to find out about.

if I have a list of "partners sites", they appear site wide; you are saying this is bad.

what about blogrolls; what about the millions of sites who do that.

honestly i am at a loss.

It's true, there are many sites that do that, but are any of them getting a real tangible benefit from these links, or rather do these links sit there doing nothing to the sites' SERPs because they're devalued?

It'd be difficult to argue objectively that link exchanging on every single page is there for anything other than SERPs and PR benefit. Do you both really love each other's sites so very much that you casting a purely editorial vote from every single page? Is each site actually highly relevant to every page that the link appears on and its presence is going to enhance the end user's experience on every single page?

It's links like these that were the first to get devalued by Google in its recent crack down on artificial "paid" links.

Millions of sites still use a meta keywords tag to help them rank better. It doesn't mean that it actually does though.

ok,

Thank You so Much JagNet to make my understanding of life so much blurrier today.

Do you know an article supporting this??

edit: by the way, I have site wide internal links in my nav; how about that?

I'll say i have had some success with sitewide recips as recently as a year ago. So if both sites have same amount of pages (since we're talking sitewides) and both sites have the potential for each page to garner PR as a reflection of the home page, then the higher PR should be a no brainer even with the additional links. What I would imagine, is that someplace on the higher PR site is a single page that is getting a (few) decent links and spreading to the rest (this could of course be the default page). With a site wide you'll find that page. I don't think building site relationships and linking is manipulative and i haven't read anything from Google saying you should nofollow links to other sites you own or other blogs in your network. A sitewide recip like this is no different so I see no reason to expect any sort of penalty (though granted the value could be dilluted).

The other benefit is that if site PR3 has 250 other sitewides like yours, the likelihood is greater that one of those sites will want to form a relationship with yours or link to yours, etc.

Like Jagnet said, it's hard to give an "all things being equal" advice, because obviously the sites are different. Which one do you like better? As a user, which one would you rather visit? If this is a recip, that means you have sitewides to them too. Which one do you want to endorse. If the answer is neither, then why do the recip? If the answer is both then why are you limiting yourself?

Internal links are completely different to external links in that respect. That said some argue that navigational links carry less influence than in-context links -- I don't know, as I've never had the inclination to set up tests to prove or disprove otherwise. My main navigation is there because it needs to be for the end user's benefit -- how much it helps SERPs is rather by-the-by.