1 of 13

NBA Shots Analysis

(2001 - Present)

Nirmay Kathuria, Nathan Rusch, Jack Anderson,

Juan De Los Santos, Wilson von Bohlen

2 of 13

Introduction

  • Objective:
    • Use data science techniques to explore and visualize NBA shot data
    • Provide valuable insights into shooting efficiency and trends for teams and players

​

3 of 13

Dataset Overview

  • Source: Kaggle - NBA Shots Dataset (2001-Present)
  • Key Features:
    • Player Name
    • Team Name
    • Shot Location (X,Y)
    • Shot Outcome (Made/Missed)
    • Game Date

​

  • Dataset Size: roughly 4 million entries

4 of 13

Methodology

Step 1: Data Preprocessing

  • Cleaned and filtered shot data for analysis
  • Extracted date out of game_id and converted to SQL Date Obj
  • Removed unused columns

Step 2: Backend Querying (SQL)

  • Extracting shots by team and date, player

Step 3: Dash-based Web Interface

  • Users can select teams/ players and view shot distributions

Step 4: Data Visualization (Plotly)

  • Interactive visualizations of shot charts

​

5 of 13

Backend Implementation

Database: MySQL

Functionality: Queries shots for a specific team/player and date range

Code:

​

​

6 of 13

Web Application

Built with Dash (Ploty, React, HTML, CSS, Flask)

Features:

  • Dropdown for selecting NBA teams/players
  • Dynamically generated shot charts for selected teams/players

Code:

​

7 of 13

Changes & Improvements

Challenges Faced:

  • Handling large datasets efficiently
  • Query performance issues in MySQL

Future Improvements:

  • Side by side player comps
  • Differences in efficiency for each shot type
  • Interactive filters ex home and away

8 of 13

Visualizations (steph curry 2016-2019)

9 of 13

Analysis

  • A high amount of shots around the 3 point line suggests that Steph plays a high scoring role on his team
  • With roughly 1500 shots per season and make rates typically higher than miss rates…
  • One can infer that steph curry plays a very niche and specialist role on his team as a 3 point shooter

10 of 13

Visualizations (lebron james 2007-2011)

11 of 13

Analysis

  • LeBron has a high concentration of shots at the rim showing that he is the team’s main power forward
  • Even though his priority is to be upfront we can see that he can be versatile and take shots were is needed like far back in the 3 point area
  • As the years go by we can see that the amount of shots missed gets surpassed by the amount of shots he makes. He surpasses the amount in 2011 and in the years prior to that we can see the ratio of missed to made shrinks slowly.

12 of 13

Conclusion

Key Takeaways:

  • Analyzed NBA shot data over multiple seasons
  • Built a web based tool for dynamic shot visualization
  • Identified trends in shot selection and shot efficiency

Next Steps:

  • Enhance database queries/structure for improved speed
  • Add output useful/insightful statistics (longest shot, most frequent shot, most frequently missed shot)

13 of 13

Q&A

Questions?