8letters
Writing
- Joined
- May 27, 2013
- Posts
- 2,160
If you don’t like discussion of story statistics, HIT BACK NOW!!!!
I’ve been collecting 30-Day Voting Top Lists data since 8/30. Caveats about the data are here.
Let’s say we have a story that has N votes on one day, and then has N+1 votes on next day. It’s possible to calculate what that one vote was for that story, as long as the number of votes is equal to or less than 100*. I decided to take advantage of that to determine what is the voting distributions for each of the categories.
Let me walk through an example of what I actually do. There was a story on 9/1 that had 99 votes and a 4.81 rating, and on 9/2 had 100 votes and a 4.80 rating. I have a table that provides the points for any combination of votes and rating, which is votes*rating (I do this to avoid problems caused by rounding). So when the story had 99 votes and a 4.81 rating, it had 476 points. When it had 100 votes and a 4.80 rating, it had 480 points. 480-476 gives that the one vote was a 4.
The “One Vote Avg” column is the average vote calculated by the percentages to the left of the column. The “Delta Avg” column is the average vote from between Day 1 and Day 31 of when a story is published, from the post I linked to earlier.
Some quick points:
* Sweeps can make a single vote look like it had a ridiculous value. One Loving Wives story got what looked like a vote of 17. Interestingly, Loving Wives was the only category with a single vote that looked less than 1 (it had two stories with 0 votes). I excluded all those out-of-range votes
* The results included in this table are for stories that ended up with between 7 and 100 votes (the 30-Day Top Lists require 6 votes to make them)
* Loving Wives is the category with the most votes but it doesn’t have a lot of single votes in this table because so many of its stories quickly get over 100 votes
If you'd like to see any of the details behind the numbers, hit me up. I respond more quickly to conversations than thread replies.
I’ve been collecting 30-Day Voting Top Lists data since 8/30. Caveats about the data are here.
Let’s say we have a story that has N votes on one day, and then has N+1 votes on next day. It’s possible to calculate what that one vote was for that story, as long as the number of votes is equal to or less than 100*. I decided to take advantage of that to determine what is the voting distributions for each of the categories.
Let me walk through an example of what I actually do. There was a story on 9/1 that had 99 votes and a 4.81 rating, and on 9/2 had 100 votes and a 4.80 rating. I have a table that provides the points for any combination of votes and rating, which is votes*rating (I do this to avoid problems caused by rounding). So when the story had 99 votes and a 4.81 rating, it had 476 points. When it had 100 votes and a 4.80 rating, it had 480 points. 480-476 gives that the one vote was a 4.
The “One Vote Avg” column is the average vote calculated by the percentages to the left of the column. The “Delta Avg” column is the average vote from between Day 1 and Day 31 of when a story is published, from the post I linked to earlier.
Some quick points:
* Sweeps can make a single vote look like it had a ridiculous value. One Loving Wives story got what looked like a vote of 17. Interestingly, Loving Wives was the only category with a single vote that looked less than 1 (it had two stories with 0 votes). I excluded all those out-of-range votes
* The results included in this table are for stories that ended up with between 7 and 100 votes (the 30-Day Top Lists require 6 votes to make them)
* Loving Wives is the category with the most votes but it doesn’t have a lot of single votes in this table because so many of its stories quickly get over 100 votes
If you'd like to see any of the details behind the numbers, hit me up. I respond more quickly to conversations than thread replies.
