I like to think that we take the unconventional approach.
Last year we had everything down to a science. We had the tools. We had the know how. We could make the numbers dance and quickly come to interesting conclusions to help us write our sparkling copy.
It would have been easy to rest on our laurels.
But it wasn’t quite perfect and we had some time on our hands this summer. So we blew it up and it was good.
There was a dark side to this tale however. This year we had to rebuild everything. But hopefully from the ashes something better arose. We have two fantastic data sites for all your statistical needs (The NBA geek and Nerd Numbers) that are miles better than what we used to have. As for the analysis bit, I’m ready to step up to the plate again. Thankfully I find that for me Revision 2 is much better than the original. Case in point, let’s talk about the MVP race.
That is a table with the typical performance of opponents by position for every team. Anyone familiar with my work will remember that I pay a lot of attention to this. Let me explain. What I’m doing is revisiting the idea of adding individual opponent adjustment to the Wins Produced model. The driver for this has been the fact that currently Wins Produced divides team defense up at the team level assuming everything is equal and homogeneous. It’s a good approximation but it’s not quite right and I have been trying for a while to get at this. Why? I don’t know call it a pet peeve.
The goal then is WP48 at the player level adjust for what the player does and what his opponent does. To do this I look at Player data for each team and mapped how each opponent did by position (this also is not quite perfect as it groups multiple players together but the more minutes played by the player the closer it’ll be) . The table above breaks it down like that. The next step is a little more involved.
What I really want to know is how much better or worse than the team average is the player’s defense based on opponent production. I can do that by making a table and converting it to points margin per game (Points over Par (PoP)). What do we learn at fist glance? Boston PGs force their opponents to give up about +1.6 points per game in production beyond the regular defensive loss. Tyson Chandler is a defensive beast (Knicks opposing PFs are -0.9 PoP and C are -1.3 PoP with Chandler playing with Melo and Stat). Now comes the fun part, let’s add this into the regular Wins Produced calculation.
That’s every player this year who’s played more than 400 Minutes with the team that they’ve played with sorted by adjusted wins produced. I’ve also graded players based on their Points over Par per 48 as follows:
Interesting on this table is Camby being traded to a Houston team that looks intriguing now. Let’s get to what we came for though:
To recap:
The King does it again (in the regular season)
-Arturo
Oh one more thing:
Have a nice weekend.
Get more great NBA analysis over at Wages of Wins Journal.