As a professional statistician, I understand the myriad flaws of Hollinger’s statistics. However, this economic crisis has been hard on economists and academics. Work is scarce. I have had to scrounge for peanuts at Cirque de Soleil to feed my children. That said, I have recently come into some money because of generous support from an anonymous donor. This has allowed me to perfect my own statistical model, which vastly improves on Hollinger’s methods. I call this statistic the Player Efficiency Rating Triangulation Plus (PERT+).
PERT+ tweaks Hollinger’s formula by adding defensive variables that John has not thought of yet. He has published the PER equation, but it is impossible to understand. Unless you have a PhD, like I do. Some people don’t publish their formulas because they are scared that people will steal them and profit. Well, I am never scared. Here is the final form of my equation:
PERT+ = [uPERT+ * (lgPace/tmPace)] * (15/lguPERT+) + 100KL
In this iteration, the variable KL is a binary variable that equals 1 if the player is Kyle Lowry and 0 if the player is not Kyle Lowry. PERT+ is easy to understand because it uses the same scale as PER, with 15 being the normalized league average. Using the new equation, here are the league leaders in PERT+ for 2008-2009:
Coincidentally, Kyle Lowry’s PERT+ this season is the highest PERT+ of all time. Also, you may be asking how Ricky Rubio’s PERT+ is so high when he hasn’t even played in the NBA. To be honest, I can’t answer that. For some reason, PERT+ loves Ricky Rubio’s flowing locks.
In closing, PERT+ should become a great evaluation tool for years to come. It also sheds insight on this year’s MVP race. It brings up this question: why isn’t Kyle Lowry getting more consideration for MVP? Sure, LeBron James is good. But he’s no Kyle Lowry.