NBA Summer League has now concluded, with a number of rookie draft picks showing their impressive potential. Last week I looked at whether individual Summer League statistics
points, rebounds, and assistswere correlated with those statistics in the NBA.
Today, I decided to take a step back and see if overall Summer League performance has any correlation with NBA performance. To do this, I looked at PER, a commonly cited player efficiency metric. While certainly not perfect, PER is generally a good measure of the performance of a specific player.
If you read the previous article, you can skip the next few paragraphs explaining the method behind finding a correlation coefficient.
A correlation can be found using a linear regression. While linear regression sounds like a fancy term, all that it entails is finding the relationship between one independent variable (Summer League PER) and one dependent variable (NBA PER).
The result of a linear regression is often a correlation coefficient. Simply put, this is one number that outlines the relationship between the two variables. The correlation coefficient is always between -1 and 1 with -1 representing a perfect negative relationship (as one variable increases, the other decreases), 0 representing no relationship at all, and 1 representing a perfect positive relationship (as one variable increases, the other decreases).
Here, I looked at all players who played at least 60 minutes during their rookie Summer League and was drafted in either the first or season round, up to 2013. Instead of looking at PER as a whole, I opted instead to focus on the Top 50 PER performances and the Bottom 50 PER performances, to see if there was any difference. The Results
The results of this study essentially confirmed what I had expected: Summer League PER has no correlation to NBA PER. For the Top 50 Summer League PER group, the correlation coefficient was 0.27. This indicates a weak relationship between Summer League PER and NBA PER.
The highest PER for a NBA Draft pick rookie that played over 60 minutes of summer league belongs to Jerryd Bayless
(36), with Myles Turner
(35) ranking second and Chase Budinger
The relationship between the Bottom 50 Summer League PER group and NBA PER was nonexistent as well, with a correlation coefficient of -0.01. This shows a lack of any type of relationship between the two.
However, despite the fact that there was no correlation between Summer League PER and NBA PER, the two groups still had some differences between them that I wanted to highlight.
1. The average PER of the Top 50 group was 14.85 (league average PER is 15)
2. The average PER of the Bottom 50 group was 11.63
3. The Top 50 group included two players who did not make it to the NBA
4. The Bottom 50 group included 15 such players
5. The Bottom 50 group did not have a single player with a career PER of 20 or above. The best PER of the Bottom 50 group was 16.7, belonging to both Thaddeus Young
and Eric Bledsoe
6. In contrast, the Top 50 group contained six players with a career PER above 20: Dwyane Wade
, Russell Westbrook
, Al Jefferson
, Brook Lopez
, Kevin Love
, and Brandon Roy
7. The Top 50 group had an additional seven players with a career PER above 16.7, which led the Bottom 50 group.
To sum up, neither a tremendous nor a terrible Summer League PER suggests the future success of any specific player. However, a terrible Summer League performance may be the first indication that a player will struggle to find his way in the NBA.
Note: Of 2016 first Round picks who have played at least 60 total minutes, the top five in PER (20-28) were Kris Dunn
, Jamal Murray
, Juan Hernangomez
, Jakob Poeltl
and Taurean Prince
, while the bottom five in PER (1-9) were Georgios Papagiannis
, Dragan Bender
, Marquese Chriss
, Timothe Luwawu
and Malachi Richardson