seocrab,
Thanks for the data. Some thoughts to make your analysis more insightful and actionable.
There are two basic problems with this data. First, single variable analyses like these are correct as long as all other factors are equal (or pretty damn close). With so many potential ranking factors, we cannot assume that all the other sites you looked at have similar on page optimization, backlink profiles, etc. Second, the top 5 sites will likely be pretty close in less competitive niches, while the big dogs (payday loans, weight loss, make money, etc.) sites will be much closer throughout the top 20.
So to begin to draw some meaningful conclusions, you would need to run the analysis for the top 20 sites across 10-20 very different but highly competitive niches and then aggregate the results across all the niches (average of all #1, #2, etc). This allows you to see trends if there are any. Using the readability as an example, suppose the aggregate shows:
#1 44
#2 43
#3 46
#4 41
#5 37
If the bottom 5 looked like this:
#16 39
#17 43
#18 51
#19 41
#20 39
we would draw different conclusions than if it looked like this:
#16 30
#17 28
#18 20
#19 34
#20 33
The same would be true with keyword density or other single variables. Thanks again and keep up the good work!!!
