-gobbledygook bypasses backlink strength?

Actually, this is exactly the anti-sandbox effect The website in question is new - but without the filters, it does not rank at all for the keyphrase. Yet with the filters it's #1.

 

So, this might mean that the Google Sandbox effect boosts the ranking of a new website by such a considerable amount that other websites' ON PAGE SEO (as tested by application of the -asdf filter) is outperformed - which is the case with TimS's webpage. Since new pages may even rank #1 under these conditions, this would mean that their SERP must be calculated by a completely different algo which, over time, is (abruptly or gradually?) exchanged against the standard one for "old" pages

 

Leo, the scenario I described is exactly the opposite - and it brings into question exactly what the "-asdf filter" actually does - at least in my opinion.

In this case, a new website does not rank at all without the filter.

Yet with the filter it's #1.

That would lead us to believe it's the best SEO optimized for that keyphrase - but as I described in detail on the previous page, it's not - the only thing it has are 3 backlinks using the exact keyphrase in the anchor text.

In my opinion, Google gathers all the anchor text used to backlink to a site - separates it into words and phrases - compares those words and phrases within that gathered anchor text - and when it finds 100% of the anchor text used to link to a site has the exact keyphrase used, it adds some enormous weight to that fact alone. And also in my opinion, the -asdf filter doesn't remove the ranking effect of anchor text. At least not in this scenario

TimS

 

stripersonline,

I have also found similar evidence.

The search has placed some pages MUCH higher than they would ever merit, given their low number of backlinks. It has even totally removed other pages completely from the search that do merit being in the regular serps!

The search disrupts too many factors of Googles Algorithm to be an accurate predictor of anything.

IMHO

Caryl

 

Sorry about the misunderstanding by TimS: "...the Google Sandbox effect boosts the ranking of a new website..." was meant, more precisely, to read "...the Google Sandbox effect boosts the ON PAGE SEO ranking of a new website..."

Thus, as TimS says, the new site does NOT rank at all without the filter, but it is able even to overtake TimS' site when compared using the filter for both sites - inspite of very poor content etc. Would be interesting to note for how long this lasts.

IMHO, this explains the behavior TimS observed in his special case AND the effect TimS and mcdar observed in general.

In conclusion, the "-asdf"-filter does what was hypothesized: it DOES filter out the influence of backlinks.

 

Totally agree. It is all speculation as to what the gibberish filter accomplishes. It certainly filters or dampens one or more factors, but thus far not a thing as to the what has been proven, or for that matter disproven. For all our sites that improve using the gibberish filter there are others that loose ground, yet we do not hear much mention of it. I have seen no true correlation between it and the so called *sandbox* hypothesis. G has historically given new sites a boost (perhaps applying a dif algo) but they tend to disappear for a while, and then if meeting the algo criteria reappear. I understand the logic if G were to employ a sandbox, and it is probably a good idea, but IMHO the current contest seems to dispell the sandbox theory, at least for unique terms. In any hypothesis one must be careful the experiments are not simply aimed at what we want or wish proved. IMO it is way too speculative to state the gibberish negates the backlinks.

 

The current contest establishes unique conditions in that it permits G to deal with an enormous number of websites of almost equal "age" - they are all playing in the same sandbox. Thus, no differentiation between old and young is possible on this base, everything in ranking is based on ON PAGE and BACKLINK factors and for these websites the sandbox effect is DEFINITELY EXCLUDED!

This means, we should investigate the nigritude-ultramariners w.r.t. influence of the gibberish filter on sites with many and on others with particularly few backlinks. That might be a useful and very enlightening procedure.

 

That is just about all there is to off-page stuff. I guess you might add submitting to directories, but this is still in the end really only linking.

I haven't jumped into this discussion before because I think we are much closer to what is really going on with the "-nonsenes" search over on the McDar test thread. Unfortunately it is all mixed in with the rest of the discussion over there, but I'm happier with the decisions reached there than in this thread.

Here is quick summary are some of the conclusions from that thread:

1. It isn't the sites that are in the sandbox it is the links. Google wants to see if the links are serious or are just going to be put up for a month and then taken down again. It also has a dampening effect on the swings in the SERPs

2. The -nonesense search -- and it doesn't matter what the nonesense is -dasf is no better than -hjkloe -owtio or -zvkxm -- is thought to defeat the filter that devalues the links. So maybe it predicts where the page will be when the link probation period is over.

The thought that it disables the links and judges the page on content, or on-page items only, doesn't really make sense to me. I think the logic that leads to that conclusion is the same logic that lead early astronomers to maintain that the Earth was the center of the universe.

The proof of the link filter theory is that if you apply this test to an older established page you will not get much if any difference in the SERP placements. This can only be explained on the bases of taking the links out of the sandox. If the test eliminated links from the calculation all together then you would continue to see significant differences in all site regardless of age and stability.

3. So the test obviously works by defeating Googles intended manipulation of the pages. So we don't know what other things it also breaks in Google's algorithms or filter. For instance on the McDar thread we are sure that it also has the effect of turning Stemming off.

 

My site has been online since 1999, not sure what you mean by "established" - it's a surf fishing website, few here would know of it, but in the surf fishing world it's certainly established. I don't know if Google surf fishes or not, but I guess that's besides the point

For a two word keyphrase without any nonsense filters, my site places #2. With the nonsense filters it places #48 for the same keyphrase. Again, I'm not sure of the specifics numerically of "established", "older" and "much if any difference", but hopefully this information can help with the link filter theory.

 

some good points there, bob.

 

Well I would call that old and established unless you had added a bunch of links just lately.

In any case your results are interesting. You are first one I've heard of who got significantly -- I'd call #2 to #48 significant -- worse results.

But as I said the search obviously breaks things as far as Google is concerned so who can tell for sure what might have caused your results.

 

That would not explain this: a site a few days old, with--so far as I know--no links at all to it at the time was unfindable (down to #350, at which patience gave out) for its keyword phrase, but with the garbage defiltering came up #3 for that phrase, a reasonably competitive one ("Washington State history museums").

That cannot be the effect of devalued links, as there were none to devalue. It could only, so far as I can see, have been on-page content.

 

I completely redesigned my site this week and it's indexed today for the first time.

The main changes are that I didn't have all my site links down the left hand side first, they're now on the right. The first thing is really an H1 tag.

My SERPs position held (actually improved 1 place)

MY -asdas etc position went from 8 to 51 (actually position 58)

Seems that it's just showing you onpage optimisation with this "filter"

 

You are clever dazzlindonna!
What you all are saying is working on my sites too.
Very valuable to way to check that on-page factors are 100%!!

 

Be that as it may - that's exactly what happens with my site: I have an established site with just 28 pages, where for all uncompetitive keywords or kw compounds the ranking is almost unchanged whether I apply the filter or not. Only for 2 extremly competitive keywords and 1 competitive compound, there is an - enormous - difference, all three rate better than #10 with the filter on, while without filter they hopelessly around 100 or just below.

On the other hand, I definitely cannot explain TimS' observation reported in the post just before, *which is just the opposite* of what I observe.

 

That's likely because I'm a surf fisherman and the first time I heard the term SEO was about 2 months ago My site was optimized for other surf fisherman long before I ever worried about Google, that's probably why it's #48 with the filter and #2 without it.

The very interesting thing - and probably a bit unusual - is that the first time I ever checked Google with regard to rank, the site was #1. So that means without me ever targeting keyphrase, Google had ranked the site at #1. That's a testament to Google in my mind cause the site was built with surf fishermen in mind - it was designed so they could find everything they needed - long before I had heard of "optimization". There's no question - Google is excellent at what Google does

Now Yahoo I will never understand - the #4 rank for that same keyphrase doesn't have that keyphrase a single time on their entire page - not in title, URL, alt tags, comments - nothing...that's weird.