Board Game Popularity Analysis
Introduction:
Finding a board game that is actually fun and worth playing can be difficult since there are so many choices out there. For this project, I decided to explore exactly what characteristics make a board game highly rated and enjoyable. The data compiled looks at variables such as minimum age, maximum age, description, year published, category, mechanic, number of media mentions, and many others that strive to provide a comprehensive review of how popular a board game is. In my project, I defined the popularity of a board game as the average rating that it received on a scale of 1-10. This variable (average_rating) is a calculation based on all of the user reviews for a specific game, which can be found on the website that the data was scraped from. While I think that there are many different factors that can impact what a person thinks of a game, for example, their age, personality, interests, critical thinking skills/ability to understand the rules of the game, etc, providing a numerical evaluation of the popularity of the game takes into account the opinions of everyone who wrote a review of the game.
Data:
I collected my data from 2 separate .csv files found at the following web addresses:
https://www.kaggle.com/mrpantherson/board-game-data
https://www.kaggle.com/mcdemarco/top-5000-board-games-at-bgg.
BoardGameGeek.com is a large board game database that ranks board games based on their ratings, number of players, age, year published, etc. Both datasets that I used camed from Kaggle.com but were originially scraped from BoardGameGeeks.com. The first data, board-game-data, includes the top 20,000 board games listed on this website while the second dataset includes the top 5,000 board games. These datasets were merged together, which resulted in a small loss of observations. However, this was necessary to ensure that the final dataset was as clean and usable for analysis as possible.
Question #1: Does the category of the board game influence the average rating that it received?
Just like tv shows, movies, and books, and other common types of entertainment have genres, board games have categories based on the type of game it is! Since the data contained hundreds of different board game categories, I chose to only analyze the top 6 categories that occurred. With the exception of city building and card games, buying a game with these categories versus without gives you a higher chance of liking the game since the average rating is higher.
Question #2: Does the year that the game was published impact the average rating that it receives?
There is a positive linear relationship between the year that the game was published and the rating that it received. However, there appear to be some games where even though they are older, they are still pretty highly rated. These ones are most likely oldies but goodies! Some games are just classics and too good not to play!
Question #3: Does the number of times that a board game is mentioned in the media influence the average rating that it receives? What about the source of media that it was mentioned in: Podcast, weblink, blog, news?
There is a positive linear relationship between the number of mentions a game receives in a podcast and the average rating that it receives. There also seems to be a positive linear relationship between the number of times a game is mentioned in a blog and the average rating it receives. Contrarily, there does not seem to be a relationship between the number of news mentions a game gets and its average rating. There is also a negative linear relationship between the number of times a game is mentioned via a weblink and the average rating it receives.
Question #5: Does the mechanic involved in the board game influence the average rating that it received?
It does not appear that there is a relationship between the skills/mechanics involved in playing a specific game and the average rating that it receives.
Question #6: Are the descriptions of board games more positive or negative in sentiment?
In order for their games to be appealing to the public, Board Game Creators must work hard to write descriptions that make their game sound fun, intriguing, and exciting. In addition to the highly scientific, very numerical data above about different characteristics that influence the popularity of the game, I have put together a secondary dataset to analyze the sentiments of board game descriptions. This dataset was created using the Top 20,000 Board Games dataset.
What a relief! Board games are supposed to be fun, so thankfully, there are more positive words than negative words in their descriptions! 7,527 of the words included were positive while 5,802 of the words were negative. Furthermore, the top 10 negative words were die, opponent, dungeon, attack, lose, enemy, struggle, dark, evil, and limited. The top 10 positive words were victory, win, wins, gain, powerful, gold, bonus, winner, top, glory, and cooperative. Given the competitive nature of board games, many of these words seem to have to do with either winning or losing.