How well do my ratings predict the IMDb ratings for Eurowesterns?

I am currently constructing my ratings page here. When I am finished, there will be over 450 films included with brief commentaries.

I find the process of rating movies fascinating. For these pages, films were rated only in relation to other films in the genre. Who cares how good Yankee is compared to Citizen Kane? On the other hand, if you like the genre and are interested in exploring it, you might want to know how good one film is relative to another in the genre.

However, this means that my ratings tend to be higher than those on IMDb. How well do my ratings predict those at IMDb? Well, I ran a quickly linear regression in Excel to see. Here are the results:

My ratings run along the x axis (bottom) and the IMDb ratings run long the y axis (top to bottom). If my ratings were a perfect predictor of IMDb ratings, if I gave a movie a ‘7’ it would also recieve a ‘7’ at IMDb and all the ratings would fall on the line. This is rarely the case in the real world real relationships tend to have noise, which means they form more of a cloud around the line. ‘r’, which ranges from 0 to 1 (0 to 100%), measures how ‘cloudy’ or ‘line-like’ the relationship between x and y is.

Well, this image demonstrates that my ratings are a very poor indicator of IMDb ratings. First, the relationship between x and y is very diffuse and ‘cloud-like.’ This is reflected in the r value, with is 0.33. This means that my rating only predicts 33% of the variation in the IMDb ratings. This is remarkably bad! There could be several reasons why:

  1. Perhaps I have awful taste in movies that does not accord with most people’s.
  2. Perhaps I am using a different criteria for rating the movies. In that case, my rating and most people’s ratings are not really measuring the same thing.
  3. Perhaps I have seen a lot more movies in the genre, and have a better or worse sense of what is good and bad.

Of these three possibilities, I prefer 2 , 3 & 4. I believe that I am using a different, better criteria that simply rating very different kinds of movie on the same, simple scale. Also, I think that I have seen more films in the genre than most people and have a better sense of what is good and bad about them. Finally, I obviously a fan of the films because I am blogging about them.

Another way to evaluate the quality of my ranking would be to compare it with the Top 20 lists at the Spaghetti Western Database (here and here). This is a list of the favorite films of informed Eurowestern fans. That comparison is for another blog . . .


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: