We’re counting down the Top 10 posts of 2011 on “Innovation Evolved”. Originally published August 10, here’s #8 on the countdown.
Nothing says summer like movie blockbusters, and nothing says blockbuster quite like trilogies and multipart movies. Last week one of my sons and I got into an argument. “Everyone knows the second movie of a trilogy is always the worst,” he said. “The Two Towers, The Temple of Doom, The Bourne Supremacy…”
To which I rebutted, “Sometimes the second movie is the best: The Wrath of Khan, The Empire Strikes Back, Mad Max 2. But I think the first movie is usually the best of the three.”
“Bet you a pair of tickets to Captain America you’re wrong!” he said.
A quick Internet search turned up Dan Meth’s “
For 19 of the 30 trilogies we examined (63%), the first movie was the best. (I win! Don’t argue with your dad, kid.) For 7 of the trilogies (23%), the second movie was the best, and for only 4 trilogies was the last movie the best (Return of the King, Toy Story 3, The Bourne Ultimatum and Revenge of the Sith).
When it comes to market research, the truth is out there.
But if the first movie does poorly, don’t expect them to complete the trilogy.
— Jeffrey
The Data
Because I’m not getting that hour of my life back.
| Rank | Movie | Average | I | II | III | Delta |
|---|---|---|---|---|---|---|
| 1 | Lord of the Rings | 8.70 | 8.7 | 8.6 | 8.8 | 0.2 |
| 2 | Star Wars (Original Trilogy) | 8.60 | 8.7 | 8.8 | 8.3 | 0.5 |
| 3 | The Godfather | 8.60 | 9.2 | 9.0 | 7.6 | 1.6 |
| 4 | Toy Story | 8.27 | 8.2 | 8.0 | 8.6 | 0.6 |
| 5 | Indiana Jones | 8.17 | 8.7 | 7.5 | 8.3 | 1.2 |
| 6 | Bourne | 7.83 | 7.8 | 7.6 | 8.1 | 0.5 |
| 7 | Terminator | 7.77 | 8.1 | 8.6 | 6.6 | 2.0 |
| 8 | Alien | 7.73 | 8.5 | 8.4 | 6.3 | 2.2 |
| 9 | Back to the Future | 7.70 | 8.4 | 7.5 | 7.2 | 1.2 |
| 10 | Evil Dead | 7.63 | 7.6 | 7.8 | 7.5 | 0.3 |
| 11 | Die Hard | 7.60 | 8.3 | 7.0 | 7.5 | 1.3 |
| 12 | Pirates of the Caribbean | 7.43 | 8.0 | 7.3 | 7.0 | 1.0 |
| 13 | The Matrix | 7.43 | 8.7 | 7.1 | 6.5 | 2.2 |
| 14 | X-Men | 7.30 | 7.4 | 7.7 | 6.8 | 0.9 |
| 15 | Spider-Man | 7.13 | 7.4 | 7.6 | 6.4 | 1.2 |
| 16 | Rocky | 7.03 | 8.1 | 6.8 | 6.2 | 1.9 |
| 17 | Star Wars (Prequel) | 7.00 | 6.4 | 6.8 | 7.8 | 1.4 |
| 18 | The Lion King | 6.87 | 8.2 | 5.9 | 6.5 | 2.3 |
| 19 | Star Trek | 6.80 | 6.2 | 7.7 | 6.5 | 1.5 |
| 20 | Mad Max | 6.80 | 6.9 | 7.6 | 5.9 | 1.7 |
| 21 | Planet of the Apes | 6.70 | 8.0 | 6.0 | 6.1 | 2.0 |
| 22 | Batman | 6.63 | 7.6 | 6.9 | 5.4 | 2.2 |
| 23 | Jurassic Park | 6.57 | 7.9 | 6.1 | 5.7 | 2.2 |
| 24 | Mission Impossible | 6.50 | 6.9 | 5.8 | 6.8 | 1.1 |
| 25 | Blade | 6.43 | 7.0 | 6.6 | 5.7 | 1.3 |
| 26 | Superman | 6.27 | 7.3 | 6.7 | 4.8 | 2.5 |
| 27 | Rambo | 6.17 | 7.5 | 5.9 | 5.1 | 2.4 |
| 28 | Jaws | 5.70 | 8.2 | 5.6 | 3.3 | 4.9 |
| 29 | RoboCop | 5.50 | 7.6 | 5.4 | 3.5 | 4.1 |
| 30 | Teenage Mutant Ninja Turtles | 5.37 | 6.3 | 5.4 | 4.4 | 1.9 |
Methodological Notes
Because I’m a geek, and because you sweat the details.
- Trilogies are ranked from #1 through #30 based on the average IMDb rating for each movie in the trilogy. The tiebreaker is whichever movie has the lowest difference between its best and worst ratings – in other words, it’s a better trilogy if the movies are more consistent, which is why the original Star Wars trilogy outdoes The Godfather.
- I’m pretty confident in the top 10, but there are a lot of movie trilogies out there – no doubt there are titles I forgot to include that outrank the Teenage Mutant Ninja Turtles. There have to be! (I am sure even Splinter would agree.) That’s why I bill this as “The Top 10 Movie Trilogies & 20 Also-Rans”. In other words, it ain’t “The Top 30 Movie Trilogies”.
- And our third point is what is a trilogy anyway? Describing this research to a friend led to a heated debate about whether a sequence of 3 movies is all that it takes to be considered a “trilogy”. If I included in it the list, I say it’s a trilogy. I don’t care that there were movies after it or that the three movies don’t hold together as a story.
- The data is from a self-reported convenience sample of people who visited IMDb and voted. They are not typical of movie watchers in general. They are not even typical of people who visit IMDb in general (seriously!). But, face it, no one is going to spend hundreds of thousands of dollars to do a random-sample survey on this subject. (Unless they are, in which case they should call me.)
- I’m mixing and matching two kinds of data. Ratings for IMDb’s Top 250 movies, which I used in my analysis, only include votes “from regular voters”, whatever that means. The non-Top 250 include irregular votes, I guess. Yes, I could have gotten better data by signing up for the IMDb Premium account. I may be crazy enough to do this analysis but I am also cheap. As my son who is taking me to see Captain America tonight will attest.
Author Notes:
Jeffrey Henning
