Tag: Vega-Lite

  • Himalayan Mountaineering

    Himalayan Mountaineering

    Week 3 of 2025 of the TidyTuesday project provided a dataset on mountaineering in the Himalayas.

    Within this data, I looked at the various missions undertaken by mountaineers between May 2020 and June 2024.
    In the timeline below, it is immediately apparent that there is some form of seasonality in the timing of the missions. While most missions were successful, bad weather and poor conditions were the main reasons for early termination.
    Feel free to explore this data a little further yourself. You can filter for a specific time period in the timeline at the bottom, or use the bar graph at the top right to filter for a specific reason why a mission was terminated. If you hover over the dots, you find more detail about the mission.

    Embedding Vega-Lite

    For mor information about this data or the TidyTuesday project in general, visit the official Github repo.

  • Movie Ratings

    Movie Ratings


    Movie ratings are often our go-to guideposts for choosing the next movie to watch. Two of the most prominent websites in this area are IMDB and Rotten Tomatoes. While both platforms aim to give us insight into a film’s quality, their methodologies and outcomes can vary significantly. In this post, we’ll dive into the different ratings on IMDB and Rotten Tomatoes (RT), explore what movies are the best according to both websites and which ones are controversial. Grab your popcorn, and let’s break down the ratings!

    Let’s start by plotting all the movies from the two different databases, with IMDB-ratings on the x-axis and RT-ratings on the y-axis. It immediately becomes clear that the data are somewhat skewed to the right. This indicates that ratings on IMDB are relatively higher, generally speaking. Another remarkable feature at first glance are the datapoints at the very top of the graph, which I will come back to later.

    Agree to agree

    Let’s first take a look at movies with similar ratings on IMDB and Rotten Tomatoes:

    These movies lie, as expected, close to the 45°-line on the plot. For this example, I defined ‘similar ratings’ as ratings with a difference maximum 5 points between IMDB and RT. For example: the left-most green dot on the graph depicts the movie ‘Phat Girlz’, with an IMDB-rating of 22 and a RT-rating of 23. You can consider this movie as the worst movie with a similar rating on IMDB and RT.
    On the other end of the spectrum four movies stand out as agreed to be good: Pulp Fiction (89 vs 94), The Dark Knight (89 vs 93), The Shawshank Redemption (92 vs 88) and Inception (91 vs 87) with ratings reflecting IMDB and RT respectively.

    Agree to disagree

    We now come to the interesting part where we look at where the ratings between IMDB and RT vary quite a bit. Specifically, the movies where the rating on IMDB is either double that of RT (yellow dots), or the movies where RT’s rating is double that of IMDB (red dots). It is immediately noticeable that the former case is much more common than the latter.

    This suggests that IMDB generally gives higher ratings to movies than RT. To check this, we look at the distribution of film ratings between the two websites.

    The graph above shows that IMDB’s distribution does indeed peak between ratings of 60 and 70 out of 100. There is a striking difference in the distribution of ratings between the two sites: IMDB follows a more normal distribution, while RT’s distribution is rather flat.
    IMDB generally has higher ratings, but we can see that the number of very high ratings (85 out of 100 or more) is higher on RT. This brings us to our final point

    10/10

    While we have just seen that IMDB generally gives higher scores to movies than RT, we can see that the number of movies with very high scores is higher on the latter, as you can also see in the below variation of the now familiar dot plot.

    The red dots represent all the movies that get a perfect score of 10 out of 10 on RT. There are 2 notable findings to add to this. The first is that, unlike RT, no film on IMDB gets a perfect score. The highest movie score on IMDB is 92 (‘The Godfather’ and ‘Shawshank Redemption’). It is also worth noting that not all films with a perfect score on RT also have a high score on IMDB. 2 of these movies do not even score 65 out of 100 on IMDB (‘Straight out of Brooklyn’ and ‘Krrish’).

    Which movie to watch?

    Which movie is best to watch tonight is obviously a matter of taste, and IMDB or RT ratings are at best an indication of quality. Nevertheless, both sites remain a valuable source of information. If you are still unsure about which film to watch tonight, you can use the chart below as a guide. Discover the title and rating on both sites by hovering over the different dots. The ‘Worst Movies’ and ‘Best Movies’ rectangles contain all movies with a rating of less than 25 and more than 75 respectively on both sites.

    Embedding Vega-Lite