Introduction
Movies have seemingly changed a lot over the years. Different variables such as movie duration, genres, actors, and more have seemingly changed a lot over the years. However, are these changes actually happening? Are they symptoms of a “boomer mentality” where, back in our day, movies seemed a lot better? I hope that I can clear up that question for all while also exploring other cool factors about the movie industry.
What Data is Being Used?
I was lucky enough to find a data set on Kaggle, titled “IMDb movies dataset” (credit to the data publisher, Ashpal Singh1525). This data set includes information on movies from 1903, to 2023 - 120 years of movie knowledge! Kaggle is a website that hosts a never-ending supply of data sets for public access. You can find a data set for just about anything, and it’s a cool website to checkout. Also, they host different kinds of data science competitions, if that’s your thing.
The data set that I will be using includes a lot of variables on the 10,000 aforementioned movies. They are as follows:
- names: The name of the movie.
- date_x: The date when the movie was released.
- score: The user rating of the movie on a scale from 0-10, with decimals included.
- genre: The genre of the movie, such as Action, Adventure, or Sci-Fi.
- overview: A brief overview of the movie.
- crew: The members of the movie crew.
- orig_title: The original title of the movie.
- status: The release status of the movie.
- orig_lang: The language in which the movie was originally released.
- budget_x: The budget of the movie in the currency of the country where it was produced.
- revenue: Revenue generated worldwide.
- country: Original release country.
NOTE: if you are interested in using the my version of the Kaggle data, click here!
How Do Most Users Rate Movies?
This distribution of user ratings says that most reviewers leave a score of around 6-7, usually. The frequency of these votes are the highest as indicated by the y-axis. What I found interesting, however, was that before the User Rating started to rise at the 3 mark, the biggest category was a 0. Keep in mind, these are not row with unavailable data. Rather, these User Ratings are, quite literally, 0. It appears as though movie-goers are more likely to give your movie a score of 0 before anything between 1 and 4.5. It can be a tough crowd!
Have Movie Critics Become Harsher
Similar to the previous question, I analyzed user ratings again. However, this time I wanted to see how User Ratings have changed over time, if at all. The results here are highly interesting. 20th century critics was a lot more kind to movies than today. Or, maybe movies have just gotten worse. While that statement is quite subjective, the results can’t lie. As we hit the 21st century, critics seemingly critiqued a lot harder. As of this year, 2023, the Average User Rating was less than 4. For 120 years of movies, this is the first time that we’ve seen the Average User Rating dip below a 6. Maybe it’s time for superheroes to swap with old westerns.
Speaking of Older Movies, Which Era Earned More Revenue?
While older movies may be rated better, they surely did not routinely earn over $1 billion in revenue. It is important to note that there were older movies that broke through into the “Billion Dollar Club.” Studios like Marvel are able to push out a ton movies every year, so companies like this are bound to earn a lot in revenue, especially with their talent pool. I was curious, however, about those movies that did earn over a billion in revenue prior to the year 2000. These are the following movies:
| names | date_x | revenue |
|---|---|---|
| Titanic | 1997-12-18 | $2,222,985,568 |
| Rathinirvedam | 1978-03-08 | $1,916,346,675 |
| The Lion King | 1994-08-25 | $1,647,733,638 |
| Oryu’s Passion: Bondage Skin | 1975-06-18 | $1,610,291,205 |
| Porno document: Toruko tokkyû bin | 1982-02-26 | $1,569,323,844 |
| The Avengers | 1998-11-05 | $1,515,100,211 |
| Beauty and the Beast | 1992-06-11 | $1,268,697,483 |
| Female Market: Imprisonment | 1986-01-18 | $1,256,345,435 |
| The Wicked Reporter 2: The Rebirth of Horserace Betting | 1994-11-19 | $1,156,945,606 |
| Hentai-shiatsu-shi: Shikiyoku no kyôen | 1971-12-22 | $1,156,945,606 |
For movies produced after the year 2000, the big movie earners are:
| names | date_x | revenue |
|---|---|---|
| Avatar | 2009-12-17 | $2,923,706,026 |
| Avengers: Endgame | 2019-04-24 | $2,794,731,755 |
| Avatar: The Way of Water | 2022-12-15 | $2,316,794,914 |
| Star Wars: The Force Awakens | 2015-12-17 | $2,068,223,624 |
| Avengers: Infinity War | 2018-04-25 | $2,048,359,754 |
| Spider-Man: No Way Home | 2021-12-16 | $1,910,048,245 |
| BTS: Permission to Dance on Stage - LA | 2022-09-08 | $1,748,017,438 |
| Franco Escamilla: Eavesdropping | 2022-10-23 | $1,714,444,954 |
| Jurassic World | 2015-06-11 | $1,669,963,641 |
| The Lion King | 2019-07-17 | $1,647,733,638 |
Movies that released after 2000 performed significantly better (in terms of revenue) than movies made prior to 2000. The revenue data is also scaled so that it accounts for inflation. Older movies still performed well, even coming close to beating up some larger, modern movies such as Avatar: Way of Water. If you are curious about what the top 5 highest grossing movies over the last 120 years, then they are:
| names | date_x | revenue |
|---|---|---|
| Avatar | 2009-12-17 | $2,923,706,026 |
| Avengers: Endgame | 2019-04-24 | $2,794,731,755 |
| Avatar: The Way of Water | 2022-12-15 | $2,316,794,914 |
| Titanic | 1997-12-18 | $2,222,985,568 |
| Louis Tomlinson: All of Those Voices | 2023-03-22 | $2,081,794,006 |
Do Higher Budget Movies Earn More Revenue?
It does appear that the higher budget a movie has, the higher revenue it will generate. This makes sense as higher budget movies will have more opportunity to incorporate better visuals, sound, and more, while smaller budget movies do not possess the same resources. Given the line of standard error, you can tell that budget and revenue are highly correlated with a high confidence rate (given by the light shadowing).
At the $200M mark, there is a steady increase between the cost of production and the revenue of the movie. Movies that require smaller budgets must rely more on talent to draw in more revenue, while the high budget movies can throw enough money at the movie to reach a wide variety of audiences.
A Little Sentiment Analysis On My Own Data
I went and scraped IMDb’s website myself to gather information on the top 250 movies (by current popularity). This data is a bit newer than the previous data set used, so it’s information should be more up-to-date. I would like to perform a word count analysis on the IMDb movies that I scraped versus how older movies wrote their descriptions. It seems as though there are more spoilers in our modern descriptions, so let’s take a look to see if that’s true.
From my scraped data, the most common used bigram (two words) in the “descriptions” column is “world war”. Following this is “war II,” which is a continuation of the previous bigram. However, it is interesting that the three next most used bigrams are city names, with “york city” (New York City) as the head of the pack. It’s somewhat of a side-note, but I found it funny that “darth vader” was in the top 10 most mentioned description items. Star Wars truly had made a deep impact on the movie industry.
Now, I’d like to compare this to the other data set from Kaggle, which features movies that are much older than the ones I scraped. This graph looks like:
For older movies (before 1990), the results are somewhat similar, but also different. “World war” wins again, but is followed by “James Bond.” This was before my time, but James Bond movies ran rampant over the Hollywood industry. “War II” came up again, and was followed by numerous city names. This followed suit with my scraped data. However, I found it interesting that “civil war” was one of the most mentioned bigrams in the movie descriptions.
Have Movies Gotten Longer?
For my last graph, I wanted to illustrate how the movie industry has shifted it’s runtime, or movie duration, over time.
As of 2020, the runtime (duration) of movies has significantly increased! The average runtime of the last few years is hovering above 160 minutes - over 2 hours! Prior, movies may have pushed 150 minutes, but did not average this amount. The longest duration “era” was right before the 1970s. This was a time where most movies were either WWII-esque or a Western. According to the graph, these movies were just under 180 minutes.
After the 70s, the duration of movies certainly decreased to around 130-140 minutes. As we move throughout the 2000s, the movie duration trend stayed fairly stagnant until around 2016. This was when Marvel really started to ramp-up their movies, and the new Star Wars trilogy released around these years as well. Both of these studious/movies have longer run-times than usual, so it’s not an unexpected uptick in duration.
Conclusion
Movies have changed a lot over the years. Duration, ratings, and more have fluctuated quite a bit over the years. As of recent years, movies have gotten substantially longer, received higher budgets, and earned higher revenue scores. Interestingly, most of these graphs show an interesting phenomenon - the 1960s/1970s and 2020s are very similar in terms of movie length and revenue. We seem to be modeling off of old movies, except modern Hollywood studios are capable of producing some truly incredible visuals, audio, and even offer immersive movie experiences. So, if you’re thinking to yourself, “Are recent movies different from my childhood?” The answer is very much so, yes.