A hobby my wife and I share is watching movies. She started using Letterboxd with some of our friends and convinced me to join. I started logging my movies in October 2022 and still use the app to keep track of my viewing history. What does my Letterboxd data say about my movie watching habits? Do the genres correlate with the time of year? What genre do I watch the most?

I created the dataset using the movie title, genre, date I watched, date released, my rating, and the Letterboxd’s rating as variables. For genre I used the genre’s assigned by Letterboxd. Each film had at least one genre, but some had as many as six. I kept track of the first three genre’s listed in order, but ultimately I only used the first genre listed. I noted the date that I logged the movie as the date watched and used Letterboxd data to determine the date the movie was released. If possible I used the date of the theatrical release. Some films premiered earlier at film festivals and others were released on streaming and had no theatrical release. For films that were released on streaming I used the date they became digitally available. There are a total of 65 movies in this dataset.

The first thing I looked at was which genre’s I tend to watch. A few things surprised me about the genre data. First in collecting the data, I realized how difficult it is to categorize a movie with a single genre. In order to make sense of the varying and oftentimes overlapping genre assignments, I interpreted the top genre to be the primary genre and so on. Since each film had at least one genre I created this visualization using the primary genre for each film and created groups for horror/thriller and action & adventure because of the overlap of the two. The second thing that surprised me was that my most watched genre was drama. I was anticipating that my top genre would be horror/thriller movies because I generally consider it my favorite genre. This visualization makes me wonder if I should reconsider that.

For as long as I can remember every October I try to watch as many horror (or thriller, crime, mystery, sci-fi, anything that can be “scary”) movies as possible. I think it sets the mood for Halloween and it’s the only time I can convince my wife to watch scary movies with me. I was expecting to see a concentration of horror/horror adjacent movies logged in the month of October across 2022 to 2025. The movie I kicked off my Letterboxd account with was The Texas Chainsaw Massacre (1974) and while I do tend to watch horror/horror adjacent movies in October, I also watch films in those genres over the course of the whole year. By far October 2025 has been my most successful “spooky season” with a total of 5 movies in horror, thriller, and sci-fi. The three preceding years I logged 1-2 movies in the month of October and October 2023 does not show any scary movies watched at all. Another interesting observation from this visualization is that for the 25 months where I have logged a film, only 6 months don’t have a comedy or drama. This makes sense when you consider that out of 65 movies 29 are in the drama or comedy genre, or approximately 44% of all films logged.

This chart looks at the rating I gave each film compared to the Letterboxd rating, which is the average of all submitted reviews across the app. I’m a fairly uncritical viewer and I tend to give most movies between 3.5-5 stars. Rating films is very subjective and I have no standardized criteria that I bring into each evaluation. I am not surprised by the films that I rated more highly than the average from Letterboxd users. To me the most interesting findings in this chart are the films that have a higher Letterboxd rating. Some of them are really close, like Rosemary’s Baby (1968) or Anatomy of a Fall (2023). These films are within tenths of a point of each other and the difference appears to be due to the way Letterboxd calculates its rating, using the averages of all it’s users rates, and how I am able to rate the film in .5 increments. The films with the most dramatic differences in my rating and the Letterboxd rating, where Letterboxd rated higher are The Lord of the Rings: The Two Towers (2002), an action movie and Chinatown (1974), a thriller.

In this visualization I was interested in exploring the amount of time in days that passed between when a movie was released and when I saw it. I decided to filter the data to only include movies released in 2024 and 2025 because 2024 was when I started using the Letterboxd app seriously. 25 out of 65 total movies in this dataset were released between January 2024 and today, or approximately 38%. The movies at the very top of the list are those with the least amount of days elapsed between the release and my viewing. Of these movies I saw 8 in theaters and the rest on streaming services.

My Letterboxd dataset relies entirely on self reporting. When I decided to use this dataset I thought I was much more committed to logging my movies in Letterboxd than I actually was. Despite using the app since October 2022, I only seriously began logging my movies in January 2024.

Aside from half hearted commitment to logging movies I think some other things can account for the gaps in my dataset. I only count movies that I intentionally sit down to watch. I will not record a movie that is playing in the background or if I was on my phone/distracted while viewing. Letterboxd does allow you to log limited series, but I have inconsistently logged them so decided to leave them out of the dataset. Initially I had also planned to include rewatches, but due to inconsistent logging of these I also chose to exclude them. Overall, I’d say that I have a more diverse taste in movies than I thought when I went into this project. It’s clear that I began logging movies more seriously in 2024 and I would like to continue to track my viewing. Working with this dataset has made me want to develop a more precise rating criteria and to continue to be disciplined in consistently logging my watched films.

I think it would be interesting to explore this dataset further by adding additional variables such as where or how I watched the film (at a theater, rented at home, streaming) and whether or not I liked the film, a feature on Letterboxd, in order to see if there are movies that I didn’t like but rated highly.