Project Idea: Haikyuu Player Stats
I’ve always been a fan of the series Haikyuu!!. It’s a hilarious, gripping and wholesome high school volleyball manga/anime by Haruichi Furudate. The anime recently finished the first half of its Season 4 earlier this year and the second half was previously planned to broadcast around this time. However, due to the COVID-19 pandemic, production has been delayed and the release date of the much awaited second half has still not been confirmed. Worse still, the manga is coming to an end in a matter of (probably two) weeks.
I am devastated. I need a distraction, something to keep me going even when the series concludes. Venturing into the fandom online I found the perfect subreddit to inspire my new data viz project: visualising player statistics of the beloved characters from the series.
^ subreddit: Haikyuu Stats Ranked By Position
The Data: Player Stats
In a thread from a year ago, redittor crabapocolypse had put together a bullet point list of the volleyball players from Haikyuu!!, grouped by player position and then ranked in descending order according to the total score of the player stats. These stats came from the manga where mangaka Furudate drew player profiles, scoring each character against 6 metrics: Power, Jumping, Stamina, Game Sense, Technique and Speed. The score for each metric can be from 1 to 5.
Below is an example of main character Hinata Shoyo. All of the stats can also be viewed in English on the Haikyuu!! Fandom Wiki page.
^ Hinata’s player profile from the manga (source)
^ Fandom wiki’s visualisation of Hinata’s player stats (source)
Why a Viz is necessary
Redittor crabapocolypse has already done us a great service by grouping and ranking the players. However, she/he tops it by interpreting the data and penning some insightful takeaways about how certain characters are more outstanding than others:
In its original bullet list form, we might never have gleamed these precious findings if not for crabapocolypse’s additional summary. That form of data representation was lengthy, made it difficult to compare players, identify outliers and required the audience to remember large amounts of information. Instead, visualising the info can help us eliminate these difficulties, present the interesting findings that crapapocolypse has highlighted and provide new ways of exploring the data!
My Game Plan
I’m pretty excited about this project. There’s tons of potential for data exploration and visualisation: players can be grouped or sorted not just by the position that they play on the court, but also by their other characteristics like their school, age and even height. We can compare each player by their aggregate score or by selecting a specific metric. Additionally, we can compute the average and median statistics for each metric of each player position or each team. I’m also imagining user interface details like using team colours and player images as icons.
But first things first, we need to establish a minimum viable product (MVP) before developing each of these features. Here are some of the first steps/goals I envision going into this MVP:
Acquire the data, transform it into a machine-readable form e.g. a csv or json file.
Create a site to place the visualisations e.g. a github repository + page
Start with one visualisation. For this, I aim to present a similar amount of information as crabapocalypse first did in their original thread. So, this viz should:
rank players by his total score
display the score for each metric
optionally specify/filter by school and player position
Expect to read about my ideas and implementation in the comings posts!
Tools of Choice
<strike> I plan to clean and generate a dataset using Python (pandas), and create the interactive viz with Javascript (d3.js + plotly.js). </strike>
Iter 1/Jul 12 update : Today I found out about Plotly’s Dash, which will allow me to use Python for site hosting + data wrangling + interactive viz! So I’ll be giving that a shot.
Upon reflection, I’m still awash with bittersweet feelings that this series is ending so soon. T_T Haikyuu!! has always had a special place in my heart and I’m sure it will remain so for a long time to come. I hope this viz project will take away some of the sadness that comes with all endings.
Do you also love Haikyuu!! and have anime recommendations for me? Or do you have feature recommendations for the viz project? Let me know in the comments ~