Statistics About How Whales Use Their Votes

trogdor (64)in #steemit • 8 years ago (edited)

In order to add to the discussion of whale voting behaviors (the good, the bad, and the ugly), I dive into the data.

TLDR: Whales overwhelming vote UP content, only downvoting about 6.4% of the time. They also give a modest amount of attention to comments sections, using 14.6% of their votes on comments.

I wanted to add something to the ongoing discussion about voting behaviors (which typically revolve around WHALE voting behaviors in particular), so I thought a good first place to start was to simply mine for data and see exactly how whales are using their votes. I used a very simple code based on Piston to sift through the votes for many of the top accounts on steemwhales.com, ignoring those whales with little or no voting history. Overall I ended up pulling data from 31 of the top whales. Out off 21,850 votes that I found, 93.6% of those were upvotes, and 14.6% were used in the comments sections.

Overall Representative Whale Voting Behaviors

Next, I wanted to break this down by whale, to see how each one uses their votes specifically. Here you can see that, as expected, @wang has a flawless upvote record, and the majority of whales use more than 90% of their votes to upvote content. As you can see @dan and @dantheman are the only accounts that use more than 20% of their votes to downvote content. I think the proper way to interpret this is that Dan takes a more aggressive approach to finding and flagging inappropriate or plagiarized content. I leave it to the reader to form further conclusions from these results. Perhaps if this post sees modest success, I will dig into the data further and try to subjectively analyze whale's downvotes to qualitatively examine when those downvotes are used "for good" or "for evil".

Percentage Of Votes That Whales Use To Upvote Content

Finally, I wanted to look at how frequently Whales use their votes in the comments sections. As @dantheman was discussing yesterday, random voting in a comments section can be very profitable for whales. This is not to say that these votes are either random or purely motivated by profit, but it may be valuable to look at the trends nonetheless.

Percentage Of Upvotes That Whales Use In The Comments Sections

As you can see, @fuzzyvest, @summon, and @berniesanders top the charts for voting in the comments section. I leave it to you to form whatever conclusions you will from this. Personally, I enjoy the occasional random Bernie vote storm in a comments section. What do you think about?

If you are interested in the code that I used, it is here:

#! /usr/bin/env python3 
from piston.steem import Steem
steem = Steem()
voters = ['ned', 'dan', 'blocktrades', 'jamesc', 'berniesanders', 'smooth', 'val-a', 'dantheman', 'tombstone', 'itsascam', 'rainman', 'steemed', 'summon', 'blackjack', 'cloop1', 'pharesim', 'steempty', 'wang', 'firstclass', 'complexring', 'fuzzyvest', 'riverhead', 'abit', 'roadscape', 'steemit200', 'xeldal', 'wackou', 'arhag', 'enki', 'clayop', 'nextgencrypto', 'kushed', 'witness.svk', 'proskynneo', 'smooth.witness', 'silversteem']
voterIndex = 0
num_posts = 4000
upvotes = []
downvotes = []
numVotesOnComments = 0
x = steem.get_account_history(voters[voterIndex], end=100000000, limit=num_posts, only_ops=["vote"])
for index, item in enumerate(x):
      myItem = item[1]
      operation = myItem['op']
      opDetails = operation[1]
      voter = opDetails['voter']
      permlink = opDetails['permlink']
      weight = opDetails['weight']
      author = opDetails['author']
      if voter == voters[voterIndex]:
            msg = "Voter: " + voter + ", Author: " + author + ", Permlink: " + permlink + ", Weight: " + str(weight)
            if weight > 0:
                  upvotes.append(msg)
            else:
                  downvotes.append(msg)
            if permlink[:3] == "re-":
                  numVotesOnComments += 1
print("Whale = " + voters[voterIndex])
print("Total Votes = " + str(len(upvotes) + len(downvotes)))
print("Upvotes = " + str(len(upvotes)))
print("Downvotes = " + str(len(downvotes)))
print("Total votes used on comments = " + str(numVotesOnComments))