Acceptable range of Duplication for spinned content?

You do not have to spin content, this is a choice. Duplicate Content Penalty does not exist. If you choose to spin content then it is best to hit for 70%+ in uniqueness..

James

70% in uniqueness? no wonder why my spunned articles seem to be "dead", they were only 40% Thanks James

Any answer you get will be speculation. I don't spin anything, but when I do manual rewrites of PLR or my own articles, I try for at least 50% uniqueness. But that's arbitrary. I don't really know how anyone could give you a definitive answer to this, assuming you're worried about how Google views it. All you could hope to do is befriend someone at Google who is privy to this and willing to divulge it.

Several years ago there were a bunch of search engine patents filed that talked about using a "k-shingle" algorithm and it has been discussed in papers written by both Google and Yahoo engineers. They are using it to determine relevancy in search results. One thing that is interesting is they use stop words (ie - common words like a, an, and, is, the...) and clean them out in the shingle creation process. I think if they use a k-shingle algorithm to check duplicate content, they are probably using stop words because it results in a faster processing time and less memory usage.

However, I think we can be sure that they do not use it to find duplicate content in general. The only step in the SERP production process that Google "may" use it is after they collect the final set of pages to consider for their 1000 results. I doubt they use it because that would still be a very heavy computational load. Remember they have to deliver those 1000 results quickly and are doing millions per minute.

So I think we can assume they are not using k-shingle algorithms in real time to detect duplicate content in SERP. The question becomes, are they using it in the background processing of their database? Who knows?

As James mentioned before, there is no detectable duplicate content penalty. To be more precise, the only time duplicate content may impact you is if it is within your site. This issue has been addressed by Jerry West. He found this is not that strong and is dependent on your title tag and meta description tag. If they are the same then Google sees the two pages within the same site as duplicates. (You think that is why they notify you of duplicates in Google Webmaster Tool?) He showed by just changing those two items and leaving body content the same, resulted in NO duplicate content treatment by Google in SERP.

Rodney

PS - The shingle algorithm is used by DupeCop and others to do their uniqueness calculations. In DupeCop's algorithm they use two shingle sizes, 1 word and 3 word shingles.