This is my third analysis of the steem blockchain with Python.
Here I present fresh data - what happened between Aug. 30 and Sept 5 - in conjunction with last week's results. This way we can make an impression of how things progressed since then.
I would like to give specials thanks to: 
Most Voted Authors - August 30 - September 5
Here's how the process goes:
1. I took ~30 high-power voters (whales) from heimindanger's website.
2. I used Python to analyze their votes. I included only the votes made on posts (ignore comment votes: if permlink[:3] != "re-").
As said, I am only analyzing high-power votes here; they carry the heaviest of weight. Were I to include all the votes from all users, the numbers and graphs may probably look different (but I assume the payout/post would not). Here are this week's results:
Number of High-Power Votes Given - Table View
Number of High-Power Votes Given - Plot
Number of High-Power Votes Received by Authors - Table View
Number of High-Power Votes Received by Authors - Plot
Historic numbers of high-power votes given:
- total high-power votes given this week: 5407
- total high-power votes given last week: 6124
- total high-power votes given two weeks ago: 6076
The interpretation might be biased because of my forgetting to include 
A few points:
- overall, whales seem to be consistent with their votes
- similar to last week, some authors have dropped from the list while others have joined it
- similar to last week, the top 5 has changed
The Code
The python code behind this analysis is in my github. Use/modify/adapt it at your own discretion.
To-do:
- optimize the statistics by splitting votes into: upvotes and downvotes. Some of the authors could be on the list due to a mix of upvotes and downvotes - as was the case of some plagiarizers. This has to be clarified.
If you have ideas for other similar analytics, please suggest them below.
To stay in touch, follow 
#programming #steemit #analytics
Cristi Vlad, Self-Experimenter and Author