In this project we will investigate an NBA Dataset and perform some Hypothesis Testing.
- Analyze and perform EDA on the Dataset
- Applying knowledge of Module 3 knowledge and skills
- Answering 3 questions
For this project we investigated the NBA statistics Dataset which we found on kaggle. We applied statistical analysis on it in order to investigate 3 questions. We applied our previous knowledge of cleaning and EDA on the Dataset and made our Hypothesis based on the Dataset.
- H0 Null hypothesis: there is no significant difference in player performance when playing Home vs Away
- H1 Alternate hypothesis: there is a significant difference in player performance when playing Home vs Away
- H0 Null hypothesis: there is no significant difference in ease for teams to make the Playoffs in the East Vs the West from the last 16 Seasons
- H1 Alternate hypothesis: it is significantly easier for teams to make the Playoffs in the East Vs the West from the last 16 Seasons
- H0 Null hypothesis: There is no significant difference in teams winning who have at least one or more players who are all stars(with a DF_SCORE > 40 in their team)
- H1 Alternate hypothesis: There is significant difference in teams winning who have at least one or more players who are all stars(with a DF_SCORE > 40 in their team)
Findings
We found that there is no significant effect on player performance whilst playing at Home Vs Away
We found that it is significantly easier for teams in the East to make it to the Playoffs Vs the teams in the West. This is due to teams in the West being better and thus the threshold to make it to the Playoffs is much higher compared to the East.
We found that there is a significant effect on team wins when they have 1 or more superstar caliber player in the team compared to teams who have n superstar caliber player
Slides presentation: https://docs.google.com/presentation/d/1QouAeuQSaU6GGXPNBSBUlftpnxKaoF3KmP-dQ00WHFI/edit#slide=id.g7635a0c7f4_1_1