With the NBA Playoffs coming to a halt and the NBA draft right around the corner I thought it would be a very cool idea to try and devise a way to predict whether or not a player will have a successful career based on his first two years in the league. More specifically, I decided to compare a player’s sophomore season (2nd season in the league) to his rookie year (1st season in the league). The hypothesis I have is that if a player experiences an increase in his Player Efficiency Rating (PER) in year 2 compared to year 1, the magnitude in change will predict how successful his career will be. To test this, I will be using Pearson Product -Movement Correlation. Lets introduce these topics:
What We Will Be Looking At
PER is John Hollinger’s all in one basketball statistic rating that attempts to summarize all of a basketball player’s contributions into one number. PER strives to measure a players per minute performance. The league average PER in the NBA is 15.00. Efficiency ratings over 30 signify MVP type of years. It is also worth mentioning that PER largely takes into account offensive stats. These stats include: field goals, free throws, 3-pointers, assists, rebounds, blocks and steals, and negative results, such as missed shots, turnovers and personal fouls.
As mentioned above, to measure the level of correlation between the their success and the difference in PER I used the Pearson Product-Moment Correlation. In statistics, the Pearson product-moment correlation coefficient ( PPMCC or PCC or Pearson’s r) is a measure of the linear correlation between two variables X and Y, giving a value between +1 and −1 inclusive, where 1 is total positive correlation, 0 is no correlation, and −1. Thus for this analysis, since we are trying to prove that the greater the positive difference of PER the more accolades a player should have, we want a number as close to 1 as possible.
The Analysis
For this analysis, I took a sample of the last 26 Rookie of the Year winners and their PER for their rookie and sophomore year seasons. I then calculated the difference between the two. As a measure for success in the league I also scraped their 1st Team, 2nd Team and 3rd Team Awards along with their MVP and All Star Appearances. I also adjusted these statistics to account for the number of years each player had in the league (used a weighted version). A sample of the data can be viewed below:
Results
As you see here, the correlations were at 0.5 and above. This means that there is a moderately positive correlation between those accolades and the difference between PER in the first and second years of a players career. As well, I added Career PER to have an additional measure of how successful a player is during their career. I added the non-weighted version of those accolades as well to see what the difference would be.
Conclusions
As the results above show there is a moderate positive correlation between the difference in the PER of a player’s rookie and sophomore year seasons. Thus, we can say that the difference in PER can be used as an indication that a player will be a very successful player in his career.
If I had more time, I would like to take a larger, more diverse sample size. As well, I would like to throw in different statistics as a measure of a player’s success in the league. Thank you guys for reading!