In the Objective Top Twenty, a Certified Fresh Is a Must…But Is It Enough?

When you review the Objective Top Twenty you’ll notice that every movie has earned a Certified Fresh designation from Rotten Tomatoes. It is a dominant factor in my rating system. It may even be too dominant.

When you review the Objective Top Twenty you’ll notice that every movie has earned a Certified Fresh designation from Rotten Tomatoes. It is a dominant factor in my rating system. It may even be too dominant.

All of the analysis that I’ve done so far suggests that a Certified Fresh designation by Rotten Tomatoes is a strong indicator of a “really like” movie. The new Objective Database that I’m working with also shows that a Certified Fresh rating results in a high likelihood that IMDB voters will rate the movie a 7 or higher.

 # of IMDB Votes IMDB Votes 7+ %
Certified Fresh               19,654,608 88.2%
Fresh                  6,144,742 75.4%
Rotten                  9,735,096 48.5%

And, as you might expect, the likelihood of a 7 or higher rating stair steps down as you move into the Fresh and Rotten groups of movies.

This exposes a flaw in my previous thinking about Rotten Tomatoes. In the past I’ve indicated that I haven’t seen a statistical relationship between the % Fresh and the likelihood of a “really like” movie. And, actually, that’s a true statement. The flaw in my thinking was that because I didn’t see it I assumed it didn’t exist.

The Certified Fresh, Fresh, and Rotten designations are primarily defined by % Fresh:

  • Certified Fresh for most movies is > 75% Fresh
  • Fresh for most movies is > 60% and < 75% Fresh
  • Rotten is < 60% Fresh

If differentiation exists for these three groups then it should exist between other % Fresh groups. For example, movies that are 95% Certified Fresh should have a greater “really like” probability than movies that are 80% Certified Fresh. I now believe that I haven’t seen the difference because there hasn’t been enough data to produce stable differences.

When I begin to marry Rotten Tomatoes data with IMDB, I also get more data. Below I’ve grouped the Certified Fresh movies into four groups based on % Fresh.

Certified Fresh:  # of IMDB Votes IMDB Rating 7+ %
100%                     966,496 90.7%
90-99%               10,170,946 89.9%
80-89%                  5,391,437 87.3%
70-79%                  3,125,729 83.5%

We might be seeing the differences you’d expect to see when the units of data get larger.

So, why is this important? If we treat all Certified Fresh movies as strong “really like” prospects, we are in effect saying that we are as likely to “really like” The Shawshank Redemption (Certified Fresh 91%, IMDB Avg. Rating 9.3) as The Mask ( Certified Fresh 77%, IMDB Avg. Rating 6.9). The “really like” model becomes a more dynamic movie pre-screening tool if it can make a Rotten Tomatoes distinction between those two movies.

I believe that the database has to get much larger before we can statistically differentiate between Certified Fresh 87% movies and Certified Fresh 85% movies. But, I think I can begin to integrate the Certified Fresh groupings I developed above to create some additional means of defining quality movies within the Certified Fresh grade.

You might just see this change in next Monday’s Objective Top Twenty.

***

In looking at this weekend’s new releases, there are no sure things but three of the movies are worth keeping an eye on. The Foreigner, the Jackie Chan action thriller, is getting good early feedback from critics and IMDB voters. I expect it to do well at the box office. Marshall, the Thurgood Marshall bio-pic starring Chadwick Boseman, has received some early Oscar buzz. It appears to be headed towards a Certified Fresh rating from Rotten Tomatoes. The movie that may sneak up on audiences is Professor Marston & the Wonder Woman. Professor Marston created the character of Wonder Woman in the 1940’s. This movie tells that story. Already 34 of 38 critics have given it a Fresh rating on Rotten Tomatoes. I would expect it to receive its Certified Fresh designation by tomorrow morning.

 

 

 

 

 

 

Author: Mad Movie Man

I love good movies. In my prior life I worked with predictive models. I've combined my love of movies with my prior experience to create a simple Bayesian probability model to help select movies that you will probably "really like".

Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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 )

Connecting to %s