Dataset
For the second Viz of the dashboard week we were tasked to hit up the Google cloud’s big query database on NCAA basketball. There were multiple tables within this database, and we were warned not to go down the rabbit hole of trying to bring through all the datasets for our visualization and
failing miserably.
Now I haven’t a first clue about basketball or any sports for that matter, so this was especially challenging for me. I spent some considerable time in trying to make sense of the columns with the provided data dictionary and settled on analysing trends within the ‘mbb_games_sr’ table.
The Data – Choosing the Right Table
There are multiple tables in the dataset, each packed with different types of info—teams, players, rankings, game stats, and more. I decided to focus my analysis on just two:
-
mbb_games_sr – this gave me the actual game-level stats
-
mbb_teams – which I used to bring in team names and matchups
I kept things simple. My goal wasn’t to model every angle but to uncover a single strong narrative.
The Hypothesis – What Wins a Game?
Once I understood how games were scored—2-pointers, 3-pointers, free throws—I started wondering:
Is there a winning strategy behind how teams choose to score?
I read that some teams rely heavily on 3-point shots, while others play it safer and go for more 2-point attempts. So I decided to explore whether teams that attempted a higher percentage of 3-point shots (say, over 35%) actually won more home games.
This idea became the core of my analysis:
Do teams that shoot more 3s tend to win more at home—or is it still the traditional 2-point strategy that dominates?
Exploration
To even begin my data exploration, I had to first understand a bit more about the game of basketball. I talked to my brother and understood a little bit about the game in general and he explained about the scoring methodologies such as three-pointers and two-pointers. I hypothesized that distinction between the two points and three points scoring could be instrumental in teams’ victory because hey, it’s a simple logic score more points faster and you win more games. My visualisation was going to explore the strategies employed by the top 20 home teams (by wins) and if they had a bias towards a particular scoring strategy.
The story
I did some quick research to understand if what I thought was something actual basketball teams followed. I.e. was there really a strategy to victory if the team attempted a greater number of 3-pt shots vs 2-pt shots. Turns out it is a viable strategy by some teams and generally if your team attempts greater than 35% of their shots as a 3-pt then it is considered as a high-point strategy.
Based on this information, I decided to do a targeted analysis of success strategies employed by the top 20 teams in their home games. If they had a 3pt attempt% greater than 35% they were employing the high point strategy and if not, they were sticking to scoring more 2 pointers to clinch the win.
As an outsider to the sport my initial guess was that surely all teams would be going for the 3pts but in reality, it was a mixed bag. In-fact more teams in the top 20 employed a 2-pt strategy to remain victorious. The complete standout performer with a high win-rate and an outstanding 45% 3-pt strategy was Villanova. However, the analysis of why and how this team was able to consistently employ a high-risk strategy to victory will have to remain for another dashboard!
