Sacramento Moneyball

Visualise data for Kings Guard Gaming aiming to suggest helpful insights

Comment

Inspiration

We followed the prompt. ‾\_(ツ)_/‾ Well, we followed it as best we could. The inspiration for the full scope of the project was only realized after we talked out our idea with a mentor. He pointed out the entrepreneurial value in the project, which was quite the confidence boost.

What it does

The project is a visualization of the strengths and weaknesses of the Kings Guard Gaming among the NBA 2K League. The system can be generalized for other teams (merely a matter of replacing certain values), but is designed specifically for the Kings to highlight the project's potential usefulness as a personal system. With it, planning winning strategies for the next season becomes far easier to visualize, and strategies can be formed to combat specific teams or even specific players.

How we built it

We used matplotlib and pandas to visualise and compare data of different teams and players. We used the Offensive Rating/Offensive Efficiency and Defensive Rating/Defensive Efficiency to generate a NBA Power Ranking Chart. This can be used as numerical representations of competitive strength, often directly comparable so that the game outcome between any two teams can be predicted. The display interface is a combination of HTML/CSS, completely homebrew, as UI was not our primary focus. :)

Challenges I ran into

Our team dropped two members, going from 4 people to 2 (one left before dinner, the other around 9PM or so), which means we lost two contributors, severely limiting our potential scope. Weighting the data is especially difficult given the small data size; I was forced to be rather generous with the definition of "defensive" maneuvers to weight a team's defensive capabilities. Offensive capabilities were largely straightforward, but you cannot win a game using offensive strategies alone.

Accomplishments that I'm proud of

Constructing templates that allowed for fast crunching of numbers is something I'm rather proud of, even if the algorithm used to weight the data isn't perfect. I primarily have experience with UI and game logic, so data processing is not my strong suit.

What I learned

"Weighting algorithms" are easy to suggest when you aren't the one being forced to decide what to weight. Also register your domain early, not for fear that it'll be taken but rather for fear that they'll make you submit a government ID and take up to 24 hours to register the domain.

What's next for 2Kings

Sleep, probably. Maybe a bagel.

Built With

Share this project:

Updates