vikkieessien / ironhack-project-1

Ironhack-project-1 - NBA

NBA 1984-2018: How the game changed in 30 years

Data Thieves Project by:

• Victoria Essien
• Caner Bulut
• Daniel Quer

Goals of the project

• As NBA fans, we have seen how the game has become more dynamic, points, point and shooting guards have become the dominators in the court.
• We wanted to see how data backed this feeling, from two angles:
• The performance angle: how many of the points are scored from beyond the 7.25 line. Are topscorers nowadays the 3- point shooters?
• The economic angle: how much better paid are topscorers vs the rest of the team and compared to the 80s?

Data - sources, problems and limitations

We used three different data sources:

• Stats.nba.com API: we obtained here data for the performance of NBA players from 1984 to 2018. Format: JSON files
• Data World: we obtained here data for the salaries of players from 1984 to 2018. Format: 2 csv files
• We eventually had to use a package provided by a Git user to solve a referential integrity problem. Format: dictionary.

Data - sources, problems and limitations

• Data proved to be very clean, but one of the datasets contained a couple of limitations:

• 1. The Dataworld dataset was incomplete for seasons 1986-87 and 1989-90. We had to remove these seasons from the scope of our exercise, as they biased conclusions on:

• Salaries

• Positions of the players in the court

• 2. Some players that the NBA stats service recognized as active (because they had at least one minute played in a game during the season) were not present in the Dataworld dataset. These players were removed when grouping players by position and salary.

Organisation

Trello Project

Project learnings

Stats API:

• session = requests.Session()
• session.headers.update({'User-Agent': 'Custom user agent'})
• response = session.get(url)
• Working on GitHub together
• Merging dataframes and data wrangling
• Creating lookup values in data frames
• Plotting and combo colors