Is there decreasing importance of height and weight in the NBA?

Newton famously said: “If I have seen further it is by standing on the shoulders of giants.”. As the NBA develops from its 1946 beginnings, It’s interesting to see how the game and its pool of talent developed.

As basketball innately is a sport that is kind to people who are tall so they are able to scoop the ball into the net easier, I want to see if this is still the case in its 60-year development.

I want to find out if height and weight are becoming less important by the years with the increasing importance in shooting and generational development of skill to a sport that once required height and weight above all else?

Limitations:

Statistics show height and weight have increased throughout the years for the average person. Height and weight depend heavily on changes in the rule of the game and what works best based on the meta(flavor of the season) of the sport. We only have data from until 95 seasons.

Loading and cleaning the data:

imported the necessary libraries and uploaded the dataset.

Added a few variables and sorted out columns.

Looked at the statistics of every column

Dropped useless columns

Checked to see if there are any null values.

More feature engineering

Before going further with the Data — I visualized the mean height and weight per NBA season year in ascending order. I quickly found that there is a small decrease each year in height and weight.

Data Visualization:

Obviously wanted to make sure there is a correlation between height and weight.

The correlation between player height and weight is 0.828893393177432. The statistical significance of this relationship is 0.0

So as predicted, not surprisingly, height and weight are two closely related variables.

Next, I made a graph clearly illustrating the small decrease of average height and weight each NBA season.

First and Second Test

I used a chi-squared test to look at the relationship between height/weight and season in the NBA

With a P-Value of 0.86, larger than the standard significance value of 0.05, we fail to reject the null hypothesis and conclude that weight is not statistically significant with the NBA season.

With a P-Value of 0.91, larger than the standard significance value of 0.05, we fail to reject the null hypothesis and conclude that height is not statistically significant with the NBA season.

Multivariate Regression:

In both, it shows that the R squared is really small so that there is something that is affecting our data. It shows that height and weight do not change dramatically each NBA season.

Conclusion:

In conclusion, the average height and weight of an NBA player does show a declining trend from the 90’s era. Though it is not substantial yet, the decrease within recent years shed light that players are getting more athletic and also more skilled — leading less reliance on physicality and more on skill. we fail to reject the null hypothesis and conclude that weight is not statistically significant with the NBA season.

However, I feel that in the upcoming decades there will be a stronger statistical significance.

  1. Nba is playing small ball
  2. The playstyle requires more skill
  3. More emphasis on referees
  4. GSW success may have started the small ball
  5. NBA Globalization improved skills for everybody.

More Data Visualization:

Some more graphs