Final Project - My Top TV Shows

Author

Ty McCaw

Introduction

For this project I wanted to see how my top TV shows stack up against each other. I wanted to use their IMDb and Rotten Tomatoes rating to see how my ordering of the shows compares to their official rankings based on their ratings. My big question is: What kind of relationship do the scores of both IMDb and Rotten Tomatoes have? I hypothesize that it is a positive relationship with a few outliers. I had a lot of excitement coming into this as I am excited to put my newly acquired skills to the test.

Setting up the Data

First I went to Kaggle and found a data-set with thousands of shows on it coming from Hulu, Disney Plus, and Netflix. This is perfect as I now have all of my top shows in one place! I imported this data and did a bit of data wrangling to narrow down the data-set and gain the additional variables I needed.

Here is where you can grab the Link: https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQDYx5yhtVtwQL9jXQHvujswAej326lpNg-Jb2nhBqxfxNw?download=1

Initial Visualizaions

I now want to wrangle the data in a way that I can use the rating data in meaningful ways. I first wanted to see how the scores of Rotten Tomato and IMDb ratings compare to each other. Are there multiple opinions on each show, or is it a more uniform consensus all around? I have kept the code here to showcase what kind of data wrangling I performed.

#grabing the data
all_tv_show_data <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQDYx5yhtVtwQL9jXQHvujswAej326lpNg-Jb2nhBqxfxNw?download=1")

all_tv_show_data <- all_tv_show_data %>% 
  mutate(
    imdb_rating = str_remove(IMDb, "/10"),
    imdb_rating = as.numeric(imdb_rating),
    imdb_rating = imdb_rating * 10,
    rotten_rating = str_remove(`Rotten Tomatoes`, "/100"),
    rotten_rating = as.numeric(rotten_rating)
  )

rating_differences <- all_tv_show_data %>%
  select(ID, Title, Year, imdb_rating, rotten_rating) %>% 
  mutate(rating_diff = abs(imdb_rating - rotten_rating)) %>% 
  arrange(desc(rating_diff))

rating_differences %>% 
  ggplot(aes(x = rating_diff)) +
  geom_histogram(binwidth = 5, fill = "blue", color = "black") +
  labs(
    title = "Distribution of Rating Differences",
    x = "Difference Between IMDb and Rotten Tomatoes",
    y = "Count"
  )

This histogram shows the distribution of the difference between IMDb ratings and Rotten Tomato scores. This shows that most of the ratings are very close to each other, but there are a few outliers that have huge gaps. Looking into the data we see that most of these shows are not well known. It is nice to know, however, than many different point of views are being put out on the internet so people have many sources to go to an make their own decisions.

This table shows the top 10 Rotten Tomato rated shows. Below I will compare this table to the top 10 rated shows from IMDB.

Title	Year	rotten_rating	imdb_rating	source
Breaking Bad	2008	100	94	Rotten Tomatoes
Rick and Morty	2013	100	92	Rotten Tomatoes
Stranger Things	2016	96	87	Rotten Tomatoes
Attack on Titan	2013	95	90	Rotten Tomatoes
Loki	2021	95	85	Rotten Tomatoes
Better Call Saul	2015	94	88	Rotten Tomatoes
The Mandalorian	2019	94	88	Rotten Tomatoes
Dark	2017	93	88	Rotten Tomatoes
Avatar: The Last Airbender	2005	93	93	Rotten Tomatoes
Peaky Blinders	2013	93	88	Rotten Tomatoes

Title	Year	rotten_rating	imdb_rating	source
The Secret World of Nature: Spain	2020	51	96	IMDb
Bluey	2018	71	96	IMDb
Malgudi Days	1987	62	95	IMDb
Breaking Bad	2008	100	94	IMDb
Alaska Animal Rescue	2020	42	94	IMDb
Avatar: The Last Airbender	2005	93	93	IMDb
Our Planet	2019	82	93	IMDb
Cosmos	2014	82	93	IMDb
Hungry Henry	2014	39	93	IMDb
Everyday Driver	2017	52	93	IMDb

Here you can see a huge difference between the two lists. You see a lot more well known shows for Rotten Tomatoes on the lists and some lesser known shows on IMDb’s list. This makes a lot of sense as Rotten Tomatoes is critics based while IMDb is audience based. This can provide a higher than expected rating for some shows on the IMDb list.

My Top 10 Shows

Now moving into my top 10 shows. I wanted to dive a bit deeper into my personal top ratings to see if I am with or against the majority. Now that I have the skills to build and uncover information I wouldn’t have been able to a few months ago, I wonder if my feelings towards these shows will change.

Title	Year	imdb_rating	rotten_rating
Breaking Bad	2008	94	100
Attack on Titan	2013	90	95
Avatar: The Last Airbender	2005	93	93
Money Heist	2017	83	90
Gilmore Girls	2000	81	84
Power Rangers Dino Fury	2021	68	52
The Office	2001	85	84
Good Luck Charlie	2010	70	65
Storage Wars	2010	63	57
Phineas and Ferb	2007	80	71

This table shows my ranking of my top 10 shows of all time. I have a good mix of nostalgia, obviously incredible shows, and some feel good shows. I now want to use this data to make visualizations to compare and contrast these shows.

Visualizations and Interpretations for My Top Shows

Now that I have the table for my top shows, I want to find out some cool information about them!

First off I want to see how the IMDb ratings compare to the Rotten Tomato Ratings. First, I’ll need to clean the ratings data to make them comparable. I can now compare, with a scatter plot, how each shows ratings differ between the two websites.

It’s cool to see that they are typically rated similarly, but you may see that most of the shows are rated slightly higher by IMDb.

I wanted to see if the ratings have gone up or down depending on the year. I’d like to see if my love for my early childhood shows is just nostalgia or a love for a masterpiece.

Based on this plot, I am happy to see that it is not just my nostalgia making me enjoy the show (with the exception of Storage Wars). It does seem that my earlier in life shows did have a better rating which makes me think that I should just keep re-watching old show instead of finding new ones. I am also being told that this is why I should watch One Piece…

Visuals and Interpretations for Avatar the Last Airbender

Now, I want to dive a bit deeper into my favorite show of all time, Avatar: The Last Airbender. At the time of this project, I am watching this show for the 5th time. I adore this show’s character development, emotion, and animation. Now that I have developed some skills to get my hands on some of its data, I thought I’d take advantage.

I went onto IMDb’s website and scraped some data from the ATLA page. I moved it into a data-frame hosted here: https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQAGXu2YYu9yR5Wv7TV5pMUaAYnXH6S4OcCtO0DuuowMUg8?download=1

Visualization one will look to see how the average ratings have changed for each season.

It Looks as though this column chart proves that this show had improved over each season. I would 100% agree with this as I believe the best season is season 3. This also shows that ATLA is one of few shows that improves as time goes on. One of the best reasons as to why this is an all time classic.

This visualization will show us the distribution of the ratings throughout the show

This shows that there is a high average rating for each episode. This should highly draw someone to watch this masterpiece as a higher rated IMDb score points to audience approval.

This visualization will show us the how to ratings have moved over time.

This shows a slight uptrend in in the ratings per episode as time went on. However it is seen that there are some less rated episodes which point to filler episodes. All in all, this visualizations proves that when watching this show there are plenty of amazing episodes from start to finish.

This visual will show what the highest rated episodes are.

episode_name	episode_date	episode_description	episode_rating	number_of_ratings	season
S3.E21 ∙ Sozin’s Comet, Part 4: Avatar Aang	2008-07-19	Aang’s moment of truth arrives. Can he defeat Ozai, the Phoenix King? Will he be forced to take a human life? All the characters face their greatest challenges.	9.9	26.0	3
S3.E20 ∙ Sozin’s Comet, Part 3: Into the Inferno	2008-07-19	Zuko battles his sister with Katara’s help for the right to be named Fire Lord. Elsewhere, Aang meets Ozai just as the comet arrives.	9.8	15.0	3
S1.E20 ∙ The Siege of the North, Part 2	2005-12-02	As the Fire Nation continues its assault on the Northern Water Tribe, Sokka, Katara and Yue set out on a search for Aang, being guided by his spirit.	9.6	8.8	1
S2.E7 ∙ Zuko Alone	2006-05-12	As Zuko tries to make it by in exile without his uncle, he remembers how his father became Firelord and what happened to his mother.	9.6	11.0	2
S2.E20 ∙ The Crossroads of Destiny	2006-12-01	The heroes work together to stop Azula’s destructive plans; Zuko finds companionship in an unlikely source and learns the time has come to make an imperative choice about his future.	9.6	9.9	2

This tells us that the higher rated episodes tend to be later in the season with the top two being the last two episodes. This table can persuade someone to make sure you get to the end because these ATLA seemed to have saved the best for last.

This visualization will show us if the number of ratings have an effect on the rating itself.

There isn’t a strong correlation to a higher rating if more people rate, but it is definitely there as two of the most rated shows are among the highest rated episodes.

Conclusion

As this project comes to a close, so does my time in this class. I have enjoyed learning about this new and sometimes confusing language. However, I am very glad that I was able to challenge myself and push to become better. Because of this, I was able to do something this cool. I get to look back at my life through TV and learn new aspects of them I couldn’t have before. I had always thought that building something like this was far out of my reach, but with a bit of hard work and tears I made it possible. Thank you for reading!

--- title: "Final Project - My Top TV Shows" author: "Ty McCaw" toc: true # Generates an automatic table of contents. format: # Options related to formatting. html: # Options related to HTML output. code-tools: TRUE # Allow the code tools option showing in the output. embed-resources: TRUE # Embeds all components into a single HTML file. execute: # Options related to the execution of code chunks. warning: FALSE # FALSE: Code chunk sarnings are hidden by default. message: FALSE # FALSE: Code chunk messages are hidden by default. echo: TRUE # TRUE: Show all code in the output. --- ## Introduction For this project I wanted to see how my top TV shows stack up against each other. I wanted to use their IMDb and Rotten Tomatoes rating to see how my ordering of the shows compares to their official rankings based on their ratings. My big question is: What kind of relationship do the scores of both IMDb and Rotten Tomatoes have? I hypothesize that it is a positive relationship with a few outliers. I had a lot of excitement coming into this as I am excited to put my newly acquired skills to the test. ## Setting up the Data First I went to Kaggle and found a data-set with thousands of shows on it coming from Hulu, Disney Plus, and Netflix. This is perfect as I now have all of my top shows in one place! I imported this data and did a bit of data wrangling to narrow down the data-set and gain the additional variables I needed. ```{r} #| label: Librarying #| echo: false #| include: FALSE library(tidyverse) # For all the tidy things! library(lubridate) # Convenient transforming of date values. library(knitr) ``` Here is where you can grab the Link: <https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQDYx5yhtVtwQL9jXQHvujswAej326lpNg-Jb2nhBqxfxNw?download=1> ## Initial Visualizaions 1. I now want to wrangle the data in a way that I can use the rating data in meaningful ways. I first wanted to see how the scores of Rotten Tomato and IMDb ratings compare to each other. Are there multiple opinions on each show, or is it a more uniform consensus all around? I have kept the code here to showcase what kind of data wrangling I performed. ```{r} #| echo: true #| label: Grabbing and Wrangling data #grabing the data all_tv_show_data <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQDYx5yhtVtwQL9jXQHvujswAej326lpNg-Jb2nhBqxfxNw?download=1") all_tv_show_data <- all_tv_show_data %>% mutate( imdb_rating = str_remove(IMDb, "/10"), imdb_rating = as.numeric(imdb_rating), imdb_rating = imdb_rating * 10, rotten_rating = str_remove(`Rotten Tomatoes`, "/100"), rotten_rating = as.numeric(rotten_rating) ) rating_differences <- all_tv_show_data %>% select(ID, Title, Year, imdb_rating, rotten_rating) %>% mutate(rating_diff = abs(imdb_rating - rotten_rating)) %>% arrange(desc(rating_diff)) rating_differences %>% ggplot(aes(x = rating_diff)) + geom_histogram(binwidth = 5, fill = "blue", color = "black") + labs( title = "Distribution of Rating Differences", x = "Difference Between IMDb and Rotten Tomatoes", y = "Count" ) ``` This histogram shows the distribution of the difference between IMDb ratings and Rotten Tomato scores. This shows that most of the ratings are very close to each other, but there are a few outliers that have huge gaps. Looking into the data we see that most of these shows are not well known. It is nice to know, however, than many different point of views are being put out on the internet so people have many sources to go to an make their own decisions. 2. This table shows the top 10 Rotten Tomato rated shows. Below I will compare this table to the top 10 rated shows from IMDB. ```{r} #| label: Rotten Tomatoes Top Shows #| echo: false rt_top <- all_tv_show_data %>% select(Title, Year, rotten_rating, imdb_rating) %>% arrange(desc(rotten_rating))%>% mutate(source = "Rotten Tomatoes") %>% head(10) kable(rt_top) ``` ```{r} #| label: IMDb Top Shows #| echo: false imdb_top <- all_tv_show_data %>% select(Title, Year, rotten_rating, imdb_rating) %>% arrange(desc(imdb_rating)) %>% mutate(source = "IMDb") %>% head(10) kable(imdb_top) ``` Here you can see a huge difference between the two lists. You see a lot more well known shows for Rotten Tomatoes on the lists and some lesser known shows on IMDb's list. This makes a lot of sense as Rotten Tomatoes is critics based while IMDb is audience based. This can provide a higher than expected rating for some shows on the IMDb list. ## My Top 10 Shows Now moving into my top 10 shows. I wanted to dive a bit deeper into my personal top ratings to see if I am with or against the majority. Now that I have the skills to build and uncover information I wouldn't have been able to a few months ago, I wonder if my feelings towards these shows will change. ```{r} #| label: My Top 10 Shows #| echo: false tys_top_shows <- all_tv_show_data %>% filter(Title %in% c("Avatar: The Last Airbender", "Attack on Titan", "Power Rangers Dino Fury", "Money Heist", "Breaking Bad", "Gilmore Girls", "Storage Wars", "Good Luck Charlie", "The Office", "Phineas and Ferb")) %>% select(Title, Year, imdb_rating, rotten_rating) library(knitr) tys_top_shows %>% kable() ``` This table shows my ranking of my top 10 shows of all time. I have a good mix of nostalgia, obviously incredible shows, and some feel good shows. I now want to use this data to make visualizations to compare and contrast these shows. ## Visualizations and Interpretations for My Top Shows Now that I have the table for my top shows, I want to find out some cool information about them! 1. First off I want to see how the IMDb ratings compare to the Rotten Tomato Ratings. First, I'll need to clean the ratings data to make them comparable. I can now compare, with a scatter plot, how each shows ratings differ between the two websites. ```{r} #| label: Comparing IMDb and Rotten Tomatoes #| echo: false tys_top_shows %>% ggplot(aes(x = imdb_rating, y = rotten_rating)) + geom_point(size = 3, color = "blue")+ geom_text(aes(label = Title), vjust = -0.5, size= 3) + labs( title = "IMDb vs Rotten Tomatoes Ratings", x = "IMDb Rating (out of 100)", y = "Rotten Tomatoes Rating (out of 100)" ) ``` It's cool to see that they are typically rated similarly, but you may see that most of the shows are rated slightly higher by IMDb. 2. I wanted to see if the ratings have gone up or down depending on the year. I'd like to see if my love for my early childhood shows is just nostalgia or a love for a masterpiece. ```{r} #| label: Ratings by Year #| echo: false tys_top_shows %>% ggplot(aes(x = Year, y = imdb_rating)) + geom_point(size = 3, color = "darkorange") + geom_text(aes(label = Title), vjust = -0.5, size = 3) + labs( title = "IMDb Ratings by Release Year", x = "Year", y = "IMDb Rating" ) ``` Based on this plot, I am happy to see that it is not just my nostalgia making me enjoy the show (with the exception of Storage Wars). It does seem that my earlier in life shows did have a better rating which makes me think that I should just keep re-watching old show instead of finding new ones. I am also being told that this is why I should watch One Piece... ## Visuals and Interpretations for Avatar the Last Airbender Now, I want to dive a bit deeper into my favorite show of all time, Avatar: The Last Airbender. At the time of this project, I am watching this show for the 5th time. I adore this show's character development, emotion, and animation. Now that I have developed some skills to get my hands on some of its data, I thought I'd take advantage. I went onto IMDb's website and scraped some data from the ATLA page. I moved it into a data-frame hosted here: <https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQAGXu2YYu9yR5Wv7TV5pMUaAYnXH6S4OcCtO0DuuowMUg8?download=1> ```{r} #| label: Reading in ATLA #| echo: false #| include: false avatar_data <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/mccawt_xavier_edu/IQAGXu2YYu9yR5Wv7TV5pMUaAYnXH6S4OcCtO0DuuowMUg8?download=1") ``` 1. Visualization one will look to see how the average ratings have changed for each season. ```{r} #| label: ATLA Visualization 1 #| echo: false avatar_data <- avatar_data %>% mutate(season = as.numeric(substr(episode_name, 2, 2)) ) season_avg <- avatar_data %>% group_by(season) %>% summarize( avg_rating = mean(episode_rating, na.rm = TRUE) ) season_avg %>% ggplot(aes(x = as.factor(season), y = avg_rating))+ geom_col( fill = "lightblue",color = "blue")+ labs( title = "Average Episode Rating by Season", x = "Season", y = "Average Rating") ``` It Looks as though this column chart proves that this show had improved over each season. I would 100% agree with this as I believe the best season is season 3. This also shows that ATLA is one of few shows that improves as time goes on. One of the best reasons as to why this is an all time classic. 2. This visualization will show us the distribution of the ratings throughout the show ```{r} #| label: ATLA Visualization 2 #| echo: false avatar_data %>% ggplot(aes(x = episode_rating)) + geom_histogram(binwidth = 0.3, color = "red", fill = "darkred") + theme_minimal() ``` This shows that there is a high average rating for each episode. This should highly draw someone to watch this masterpiece as a higher rated IMDb score points to audience approval. 3. This visualization will show us the how to ratings have moved over time. ```{r} #| label: ATLA Visualization 3 #| echo: false avatar_data %>% ggplot(aes(x = episode_date, y = episode_rating)) + geom_point(color = "darkgreen") + theme_minimal() ``` This shows a slight uptrend in in the ratings per episode as time went on. However it is seen that there are some less rated episodes which point to filler episodes. All in all, this visualizations proves that when watching this show there are plenty of amazing episodes from start to finish. 4. This visual will show what the highest rated episodes are. ```{r} #| label: ATLA Visualizaton 4 #| echo: false avatar_data %>% arrange(desc(episode_rating)) %>% head(5) %>% kable() ``` This tells us that the higher rated episodes tend to be later in the season with the top two being the last two episodes. This table can persuade someone to make sure you get to the end because these ATLA seemed to have saved the best for last. 5. This visualization will show us if the number of ratings have an effect on the rating itself. ```{r} #| label: ATLA Visualization 5 #| echo: false avatar_data %>% ggplot(aes(x = number_of_ratings, y = episode_rating)) + geom_point() + geom_smooth(method = "lm") + labs( title = "Do More Ratings Lead to Higher Scores?", x = "Number of Ratings (in thousands)", y = "Episode Rating" ) ``` There isn't a strong correlation to a higher rating if more people rate, but it is definitely there as two of the most rated shows are among the highest rated episodes. ## Conclusion As this project comes to a close, so does my time in this class. I have enjoyed learning about this new and sometimes confusing language. However, I am very glad that I was able to challenge myself and push to become better. Because of this, I was able to do something this cool. I get to look back at my life through TV and learn new aspects of them I couldn't have before. I had always thought that building something like this was far out of my reach, but with a bit of hard work and tears I made it possible. Thank you for reading!