Stop Gamb(al)ling

• 

  The final graphic displayed after analyzing and crunching NBA player stats.

Inspiration

We were inspired to create this project after a team member's uncle had become financially incapable to support his family after losing a significant amount of money gambling on the NBA finals.

What it does

This program was created to raise awareness of the extremely low odds of successfully placing bets on NBA games.

How we built it

This program scrapes the NBA API for various player statistics using regular expressions and python. It calculates a player effectiveness rating (PER) for each player based on these statistics and the average for statistics in the NBA. Next, the program uses a machine-learning algorithm to determine any outlier effectiveness ratings based on other team members' effectiveness rating, games played, and minutes played and filter out outlying ratings of players who have not played for the majority of the season. Then, the program computes average effectiveness ratings for each team based on the new filtered list of effectiveness ratings. Finally, we used java graphics to display charts of average team effectiveness compared to the highest player effectiveness rating.

Challenges we ran into

One of the biggest challenges was finding a database to collect all the players in the NBA and their stats for the 2017-2018 season. We were eventually able to find the NBA API and the basketball-reference API. We combined the player-id keys from basketball-reference with the NBA API to find player statistics.

Accomplishments that we're proud of

We are proud of the contribution we are making to raise awareness about the poor odds of gambling, one of the major widely undiagnosed epidemics in modern society.

We are also proud to have picked up various algorithms and programming techniques while developing this project.

What we learned

We learned a lot more about parsing large amounts of online data and filtering for specific pieces of data. We also learned a bit more about developing machine-learning algorithms and translating data into a graphic.

What's next for Stop Gamb(al)ling

Next, we are looking to process and analyze data from previous NBA seasons and display graphics of the odds of successful betting rates in past years. We are also looking to refine our algorithm to filter out players who did not have a significant impact on teams' successes in different seasons. We are also looking to add a calculator to determine the odds of placing successful bets when given the two teams playing.

Built With

• graphics
• java
• python
• regex