What were the original ratings of the Top 250 by voters? Find out here.

  • 3
  • Idea
  • Updated 6 years ago
Archived and Closed

This conversation is no longer open for comments or replies and is no longer visible to community members. The community moderator provided the following reason for archiving: Old thread

Your weighted ranking in your Top 250 list is not a straightforward reporting of readers' rankings, but one that has been adjusted using Bayesian statistics. However the equation used does allow one to recover the original rankings. Out of curiosity, I did that. Five of the original top 10 remained in the new top 10 (including the present #s 1,2,3,5,and 8,) However, a new set of movies arrived in the top 10, including two Chaplins (City LIghts and The Kid), a Keaton (The General), a Marlene Dietrich (Witness for the Prosecution) and a Jimmy Stewart (Best Years of Our LIves). Four of these 5 were not even in the top 100 using your Bayesian formula. I think I and perhaps others would rank the newcomers as high as or higher than many presently ranked higher on your list. Perhaps there is great justification for the Bayesian approach, but if not, it might be interesting to let readers see the original rankings. The new top 250 are here.
https://dl.dropboxusercontent.com/u/7...
Photo of Lance Wallace

Lance Wallace

  • 2 Posts
  • 1 Reply Like

Posted 6 years ago

  • 3
Photo of bluesmanSF

bluesmanSF, Champion

  • 10815 Posts
  • 6429 Reply Likes
Since only votes from regular voters are counted, for the top 250, and since we're not told what that means, you count is probably somewhere in between. Presumably, the regular voters issue discounts votes from ballot-stuffers (which you list is not doing).

Interesting, though.
Photo of Dan Dassow

Dan Dassow, Champion

  • 13515 Posts
  • 13876 Reply Likes
Interesting! The math checks out. This is significantly different than:
Highest Rated Feature Films With 25,000-1,000,000 Votes
http://www.imdb.com/search/title?coun...
Photo of Lance Wallace

Lance Wallace

  • 2 Posts
  • 1 Reply Like
@bluesmanSF--

The equation provided by IDMB in their Top 250 relates their weighted rating (WR) to an original rating (R). I have assumed that their value of R has already been limited to just the regular voters. In fact, this must be the case, because otherwise the weighted rating would be counting those non-regular votes.
Photo of Dan Dassow

Dan Dassow, Champion

  • 13515 Posts
  • 13876 Reply Likes
Lance Wallace,

If the IMDb sets the value of v to all votes, then your calculation for R is correct. However, the IMDB states: "for the Top 250, only votes from regular voters are considered". In which case the value for v is likely to be smaller than what is reported and the calculation for R has a degree of uncertainty.

Regardless, the ranking based on your calculation of R (assuming v is set to all votes) is interesting. One could infer from this list that older films, such as Witness for the Prosecution (1957), are more likely have most of their votes from regular users than newer films.

This conversation is no longer open for comments or replies.